7

Materials Preparation and Handling

Probably no single element of incineration systems causes more problems than those
related to the handling of wastes. One of the most significant differences between liquid
and gaseous waste incineration systems and those for sludges and solids concerns the
equipment used to collect, transport, store, reclaim, and fire the wastes. In many cases,
weaknesses or failures in the design of materials handling subsystems have greatly reduced
the utility and increased the operating and maintenance problems and costs of solid or
sludge waste management facilities.

I.
A.

SOLID WASTES
General

In comparison with sludges, most solid wastes usually have a low bulk density and, except
for process wastes, include a substantial fraction of cellulose (wood, paper, cardboard).
The low bulk density requires relatively large storage space and often involves rehandling
of the waste (stacking) to increase storage capacity. For hazardous waste materials that
present a toxicity problem to plant personnel, enclosed storage such as single-outlet bins,
multiple-outlet silos, and portable bins are required. Municipal solid wastes can be stored
using a number of techniques, as shown in Table 1.
Multiple-outlet silos are useful to store small or medium quantities since they rely on
gravity to discharge the solids. Therefore, hopper slopes and outlet dimensions must be
generous enough to ensure uninhibited flow. Single-outlet bins are common in industry.
These bins use funnel-flow with steep, pyramidal hoppers sloping 60 or less from the
horizontal. Portable bins are commonly used for on-site waste accumulation at industrial
and commercial sites. In many instances, the container is equipped with a hydraulic
compactor to increase capacity and reduce the frequency of pick-up.
In small facilities, domestic solid wastes are often dumped on a tipping floor and
pushed about or stacked using a front-end loader. In larger plants, a concrete pit and bridge
crane system is more common. Pit and crane systems are much more costly than the floor
dump approach.

Table 1 Types and Characteristics of Dry Bulk Storage

Storage technique
and method of
reclaim=discharge
Stockpiles
Bottom tunnel
Bucket wheel
Scraper truck
Front-end loader
Multiple-outlet silos
Mass flow
Expanded flow
Funnel flow
Single-outlet bins
Mass flow
Expanded flow
Funnel Flow
Portable bins
Funnel flow
Mass flow
Concrete pits
Grapple

Storage capacity
Small (<550 m3

Medium

Large (>20;000 m3

X
X
X

X
X
X

X
X
X

X
X
X

X
X

X
X

Source: From (501).

Rubber belt and steel apron or vibrating conveyors have been used to transport raw
and shredded solid waste for short distances. Pneumatic systems have been used with
shredded wastes for transport for over 300 m. Frequent and severe problems with abrasion
wastage of the piping and with plugging indicate that pneumatic conveyors should be
selected only when absolutely necessary and designed with an absolute minimum of turns.
Conveyors of any type for solid wastes can be counted on to present a continuing problem
with maintenance and housekeeping.
Storage and reclaim of solid wastes is also a problem area. Frequent bridging
problems in chutes or in bins, compaction on standing, and other operating difficulties in
confined flow situations are common. These problems result from the tendency of solid
wastes to interlock, catch on minute projections, and compress. Forced expulsion through
flaring cross-sections (e.g., with a ram) or open storage in pits or on open floors are
preferred. For bunkered shredded waste, bins with vertical screws (to keep the bed
uncompacted) and live bottom design have been effective.
The most common materials handling options and problem areas in solid waste
management facilities are summarized in Table 2.
B.

Pit and Crane Handling of Solid Wastes

The pit and crane approach for refuse receiving, storing, and charging is used worldwide in
large incinerator plants. The crane operator is a key member of the incinerator staff. His or
her duties include diversion of troublesome bulky items; mixing of segregated, highly
combustible and noncombustible wastes to even out the heat release rate in the furnace;

Table 2 General Practices in Materials Handling for Domestic Solid Waste

Step

Small plant

Large plant

Receipt
Storage

Manual scale
Floor dump

Automatic scale
Pit=crane

Reclaim

Front-end loader
from storage
Ram
Hearth
Water quench,
drag conveyer

Grapple

Feed
Support during burn
Residue

Ram, chute
Metal grate
Water quench,
hydraulic ram
and vibrating
conveyer

Problem areas
Delays
Cleanout, vectors,
fires, mixing
Personnel hazard
Outage, jams
Overheating, clinker
Jams

Special processes
Shredding

Impractical

Ferrous separation

Battery removal

Floor dump and

visual check

Hammermill,
shear shredder
Belt or drum
magnet
Floor dump and
visual check

Explosions
Contamination
Delays

and relocating and stacking the pile of waste that builds directly below the tipping
positions in order to increase the usable storage capacity of the pit.
In most plants, the pit is sized to hold waste corresponding to three days production.
The pit is usually constructed to match the ultimate capacity in plants laid out with
expansion in mind. The type of construction and the cost of outages for expansion of the
pit at a later time make this a cost-effective investment decision. The usable volume of the
pit certainly includes all volume below the tipping floor. By stacking the waste against the
back wall, about 50% of the pit volume above the tipping floor level can also be assumed
to be available.
The pit is approximately 9 to 14 m deep (from the tipping floor to the bottom). This
allows several vehicles to dump before crane action to restack the waste is necessary. The
pit length is determined by the space for the feed hoppers and the furnace boiler spacing.
In some instances, the number of available tipping bays is specified. A rule of thumb (361)
for the required number of tipping bays is given by
number of bays 1=6 $ maximum number of trucks per hour

This allows about 10 minutes average for truck positioning, dumping, cleanup, and
departure. Data on the intensity of truck arrivals may be extracted from records in existing
facilities (such as the landfill). A conservative assumption suggests that the plants rated
capacity in solid waste will be delivered in 5-tonne packer trucks, delivering 7 days worth
of plant capacity in a 5-day period, and that one-half of this delivery will take place in two
rush-hour periods per day, each of 90 minutes duration. Back-calculating from this
scenario, one tipping bay is needed for each 125 tonnes of capacity. Tipping bays should
be not less than 410 cm wide. Year-long tests at three large mass burn plants in the United

States (361) indicate that an average refuse density of 325 kg=m3 should be used to
calculate both the above- and below-tipping-floor volume. Empirical estimates of the inplace density of settled refuse after approximately two days storage showed a logarithmic
relationship to bed depth. The relationship describing the average density rrefuse for
domestic, U.S. refuse is given in Eq. (2) for bunkers loaded to a uniform depth D1 . For a
bunker with a pile sloped from a front wall depth D1 to the back wall at D2 , the
relationship is given in Eq. (3).
ln D1
250 kg=m3
0:019721
50:734D2 ln D2 1:1881 ' D1 ln D1 1:1881(

kg=m3
D2 ' D1

rrefuse

rrefuse

In plants built before 1980, many of the cranes used in the United States were controlled
by an operator located in an air-conditioned cab attached to the crane bridge. The operator
is located in a stationary pulpit in more recent designs. The pulpit is located on the front
side of the pit over the tipping floor, at the end of the pit, or on the back side of the pit
between the furnace hoppers in order to provide a view of the tipping area, the storage pit,
and the furnace hoppers. In larger plants, the pulpit is sized for two operator stations. The
tipping area is very dusty, and care must be given in the design of the environmental
systems (air conditioning, fresh air make-up, etc.) to maintain a positive pressure within
the pulpit work space.
In most plants, combustion air is drawn from the pit area both for dust control and to
minimize the emission of odors. To minimize clogging of inlet screens by airborne dust,
fiber, and paper, the air intakes should be located as high as possible above the pit. The
capture velocity of the air intake openings should be in the range of 2.5 to 5.0 m=sec
(preferably closer to the lower value).
Motor controls for remote-operated refuse handling cranes are, preferably, located
off-board the crane in a clean, well-ventilated room (360). This eliminates vibrations,
shock, and impact loads and reduces maintenance difficulty and cost. With such an
arrangement, conductors are required to deliver current to the individual motors of each
motion located on the crane and to control other devices such as limit switches located on
the crane. The current-carrying conductors are high-voltage conductors and consist of
between 6 and 10 conductors per motion. The control conductors are low voltage and use 2
to 6 conductors per control device. Most often, the pulpit and control center are linked to
the crane using festooned conductor systems rather than a rigid-type conductor system
with sliding shoe-type collectors.
In many modern plants, a number of the crane motions have been reduced to
automated sequences called by the operator. These include hoisting of the loaded grapple,
homing movements of the trolley and bridge to bring the loaded grapple to a selected
hopper, lowering of the loaded grapple into the hopper, discharging the refuse, hoisting the
empty grapple, and returning to a selected pit position.
The changing nature of refuse size consist, density, and mechanical properties
demands much of the equipment used to grasp, lift, and drop waste. The types of grabs
used are most conveniently grouped by the actuation method and basic shape of the device.
Opening and closing are either by a wire rope system or an electric-hydraulic (EH) system.
The principal shapes are the two-jaw clamshell or the multi-tine orange peel type.
The long tine, clamshell grapple was preferred in United States practice for many
years. The grapple consists of two cutaway jaws with long, pitchfork-type tines fastened to

the bottom lip of the jaws. The better digging capability of this design led it to replace
conventional clamshell buckets with short teeth. Most still in use are of the three-rope
suspension (two holding ropes and one closing rope) or four-rope suspension (two holding
and two closing ropes). They are suspended from the overhead traveling crane having both
a holding and a closing hoist. Some plants use EH actuation. Although the capital cost is
higher, there is a partially off-setting benefit in increased on-line availability and rope costs
(360) because of the superior 3- to 6-month mean time before failure of the hoisting lines
(found in both rope-operated and EH devices) in comparison to the 1- to 3-month life of
the closing lines (only on the rope-operated grapples).
The orange peel grapple consists of 6 to 8 tines evenly spaced radially. Ropeactuated orange peels are of the four-rope type using an overhead crane with both holding
and closing hoists. The EH embodiment has each claw operated by an individual hydraulic
ram or a central ram with very rugged linkages to the tines. The capacity of the device is
reported to be 15% to 20% higher than the clamshell grapple. Further, the clamshell
grapples themselves are approximately 40% to 50% heavier than equivalent orange peel
systems. This increased dead weight decreases the refuse lifting capacity by 25% to 35%
in comparison to that of the orange peels (360).
C.

Size Reduction of Municipal Solid Wastes

Size reduction is a necessary step in the production of refuse-derived fuel (RDF). Also,
limited breakdown of as-received waste is sometimes used ahead of mass burning systems
to facilitate precombustion materials recovery. The characteristic dimension of refuse after
processing for size reduction is often referred to as the top-size. Clearly, both raw and
processed waste is found in a range of particle sizes: from fine dust, with dimensions
expressed in microns, to mattresses and furniture dimensioned in meters. The top size,
as most often used, refers to the particle dimension such that only 10% of the mass of the
material exceeds the top size. Note that the top size refers to the longest dimension of any
given waste element.
Size reduction for solid wastes is a complex process. Unlike coal and most mineral
material, only a small fraction of municipal waste breaks down under crushing forces (e.g.,
glass). Most of the materials in waste deform, stretch, or are simply compressed by
crushing forces. Not only is the material not subdivided, but considerable energy is
absorbed. This increases the cost of processing without providing the benefit. Thus, we
must add shearing, ripping, and cutting action.
In the late 1970s, the U.S. EPA sponsored several studies of size reduction (284,
285). The results of these studies are most interesting and useful. However, one must
exercise care in that the structured experiments and limited test objectives that guided the
research work may suggest design principles that yield marginal facilities in practice.
Refuse is inconsistent in its characteristics. Sudden shifts in waste properties that places
unanticipated, heavy demands on undersized equipment operating at high speed with
considerable rotating energy can do damage or otherwise shut down critical processes.
Four classes of equipment are commonly used for MSW size reduction:
Swing hammer-type hammermills (horizontal or vertical shaft)
Vertical ring grinder shredders
Flail mill (a lightweight swing hammermill for very coarse reduction)
Rotary shear (a slow-turning cutting-type device with circular cutter disks)

1.

Horizontal Shaft Hammermill Shredders

Horizontal shaft hammermills are the most common size-reduction devices used in refuse
applications. By adjustment of the spacing in the grate bars at the discharge, the particle
size can be varied from very coarse (say, 10 to 15 cm) to fine (2 to 3 cm). The equipment is
very rugged, and little presorting is needed. The three basic types of horizontal shaft
shredders are shown in Fig. 1. In the early 1960s and 1970s, U.S. practice favored hammer
designs involving periodic retipping of the hammers with special weld rod material.
Improved materials and increasing labor costs have shifted the preference of many plants
to a no-weld, discard concept using a hammer with a replaceable wear cap with a Brinnel
hardness over 500.
The topfeed, single-direction configuration is appropriate as the primary (first pass)
size reduction device to bring residential or light commercial refuse to a 5- to 15-cm top
size. In these devices, the in-feed is generally a free fall down a chute inclined between 60
and vertical. These devices are effective for a wide range of feed characteristics (paper,
cans, glass containers, clothing, brush and smaller branches). Disadvantages include their
tendency to reject or throw back feed materials (requiring effective containment hoods and
curtains), high shock loading on equipment, and surging output rates due to uneven feed
rates.
Also, these devices inherently generate windage, an air flow engendered by the
rotating hammers. Windage aggravates housekeeping and can be a matter of concern when
materials are shredded that present an inhalation hazard (pharmaceuticals, toxic materials,
etc.). Importantly, windage also contributes to the severity of explosions occasioned by
shredding containers filled with volatile, combustible liquids. The explosion problem is
serious and has taken the lives of workers and caused severe damage to facilities in many
plants. Explosion control using fast-acting halon (or equivalent) explosion suppression
devices has shown to be effective.

Figure 1 Horizontal shaft shredder types.

Secondary shredders, producing a 2- to 5-cm product, can be of the single-direction

or reversible center feed design. The feed to the center feed devices is a free fall close to
the rotor center line. The center feed concept was developed on the hypothesis that with
reversible motor operation, hammer wear is more even and hammer replacement is less
frequent. In fact, most shredding facilities have considerable time available for hammer
rotation, and the downtime needed to turn the hammers is not significant. However, the
reversible system presents several problems: increased tendency for rejection of feed (due
to the more open feed throat) and increased rotor windage.
Heavier, bulky trash or demolition wastes are processed best in the controlled feed
configuration where the compression feed device provides positive feed and more uniform
power load. The feed enters down a chute at an angle of 45 or more, is gripped by a
crawler apron feeder or a drum roller, and is forced at a steady rate into the hammer circle.
This design has minimal rejection and shrapnel back-fling, even power consumption, and
lesser rotor windage due to the constricted feed throat opening.
Analysis of horizontal shaft shredder data (355) showed that power consumption
varied with product particle size according to the following:
Eow 17:91 X0'0:90

r2 0:87

'0:81
Eow 35:55 X90

r2 0:89

where Eow is the specific energy consumption (kilowatt hours per metric ton processed) on
a wet waste basis as a function of X0 and X90 : the product sizes (in cm) corresponding to
the 63.2% and 90% cumulative passing, respectively. An effect of the mass of the material
being held up in the mill was also seen to increase power consumption. No consistent
pattern of power consumption with refuse mean moisture content was found, although a
lubricating effect from the moisture was expected. The power relationships were
developed from facilities handling from 4 to 90 metric tons per hour. They are, however,
generalizations and variations should be expected with design and feed differences.
Typically, motor sizes for shredding are estimated as a combination of the net power
for shredding plus the idle power (freewheeling power). Motor size can be estimated from
the power demand per ton from Eqs. (4) or (5), the desired throughput, and then division
by a factor of 0.9 kwh=hp. Clearly, a considerable safety margin is appropriate. Table 3
provides a selection chart for both shredders and motors.
Hammer wear (and consequent maintenance expense in either replacing or retipping
hammers) is second only to power in its importance in shredding economics. Analysis of
hammer wear test data (355) showed a range of 0.013 to 0.106 kg=tonne. The wide range
reflects the effects of hammer material and hardfacing alloys tested. The greatest wear was
sustained by nonhardfaced manganese steels (with a hardness as-cast of 14 Rc ). The
limiting factor for very hard alloys is the chipping of the welds under high-impact loads.
Optimum alloy hardness appears to lie in the range 48 ) Rc ) 56, with the upper value
limited by chipping tendencies. In general, this Rockwell range resulted in specific weight
loss between 0.015 to 0.03 kg=tonne.
2.

Vertical Shaft Mills

Vertical shaft mills use either swing hammers or ring grinders to reduce refuse. Larger
mills use a top-mounted, direct drive motor. Most such mills incorporate a conical shell
with a progressively smaller clearance between the shell and the rotating grinding units as
one moves toward the base. In the vertical shaft, swing hammer device, the in-feed

Table 3 Selection Chart for Shredders and Motors

Shredder size (cm)
(Hammer swing
diam.$feed width)

106 $ 150

150 $ 150

150 $ 200

180 $ 252

216 $ 252

Processing rate (T=hr)

Packer truck waste
(residential)
Bulky waste
(residential=commercial)
Motor size (hp)
(for 10-cm nominal top size)

22.7

36.4

54.5

90.0

136.4

18.2

22.7

40.9

54.5

600

800

1250

2000

3000

Source: From (461).

penetrates the hammers in a converging, conical section reaching a minimum clearance at

the neck. Reduction is by attrition between the hammer tips and anvils spaced around
the inner shell. The discharge involves a 90 turn and, due to the rotating action of the mill,
the leaving stream tends to load up one side of the discharge conveyor.
Studies of a 746-kW (1000-hp) vertical shaft machine (363) indicated that the
RosinRammler equation was adequate as a descriptor of the particle-size distribution. The
equation was originally developed in studies of coal grinding (364) and is stated as
Y 1 ' exp'x=x0 (n

6a

where
Y cumulative fraction of particles by weight less than size x
n a constant
x0 a constant known as the characteristic size defined as the size at which
63.2% of the particles are smaller
Rearranging, one notes that
!n
!
1
x
ln

1'Y
x0

6b

Thus, plotting ln1=1 ' Y ( vs. x on log-log paper yields the slope n and the
characteristic size at ln1=1 ' Y ( 1. The shredder data showed the average
x0 30 mm (range from 23 to 38 mm) and the average slope n 0:7 (range from 0.6
to 0.9). There was no observable trend in either x0 or n over the testing period (12,600
metric tons). There was no observable effect of average waste moisture content on the
particle-size analysis.
Power consumption averaged 5.71 kwh=tonne comprised of 0.08 kwh=tonne of
freewheeling power and 5.79 kwh=tonne for the shredding energy. This measure of
freewheeling power is much lower than the 10% estimated by other researchers for
horizontal shaft mills (355). Hammer wear averaged 4.74 kg=100 tonnes processed and
was approximately linear with the quantity of refuse processed.

The ring grinder mills rotate at about one-half the speed of the vertical shaft
hammermills (say, 300 vs. 600 rpm) but are otherwise similar. The rotor of these devices is
commonly driven from the bottom via a bull gear and pinion.
3.

Flail Mills

A flail mill is, in general, a horizontal shaft hammermill with lightweight hammers. The
primary function of the mill is for coarse bag breaking and limited size reduction. As such,
flail mills have considerably lower horsepower, windage, and space requirements and can
process large quantities of material at low cost. Their operation does not produce a highly
controllable size consist. Also, their low power makes them more vulnerable to interruptions unless care is given to presorting.
4.

Rotary Shears

Rotary shears are low-speed devices (say, 20 to 40 rpm) characterized by two parallel
counterrotating shafts (turning toward one another). Disks incorporating one to six cutter
teeth (two or three teeth per disk are common) are mounted on the shafts using keys. The
shafts rotate at different speeds, often about 2 : 1. Size reduction involves a shearing or
tearing action rather than the impact or crushing reduction common in the hammermill.
The differences in rotational speed and reduction method lead to a much lower power
consumption for waste processing (generally 10% to 20% of the power consumed by a
hammermill performing the same reduction). One or more pairs of shredding shafts are
mounted at the bottom of a simple, rectangular feed hopper. They discharge to a pan or
apron conveyor mounted below.
Shear shredders are particularly well suited to the processing of pallets, drums,
automobile tires, furniture, demolition waste, and other relatively large items where coarse
size reduction is sought (289). Because of their low speed (few sparks and little heating of
the material being processed), open construction, and negligible windage, they are not
prone to explosions. Also, in comparison to the hammermills, they are a low-noise and
low-dust potential device. They are less applicable to processing paper or ductile sheet
materials, especially where small or controlled particle sizes are sought.
5.

Noise Levels During Waste Processing

Many of the devices used in waste processing are inherently noisy. This presents problems
to neighboring land users and, in the plant, may present hazards to employees. Measured
noise levels associated with waste processing are shown in Table 4.
D.

Conveying of Solid Wastes

For many bulk solids, the physical and descriptive properties affecting conveying and other
materials handling operations fall within a rather narrow band. Those properties fall
outside the normal range for mixed solid waste, for processed refuse-derived fuel
(RDF), and for many waste components. Beyond their unusual average characteristics,
refuse and refuse component properties often vary widely. Generalizations concerning
materials handling, therefore, must be made with care. Further, such generalizations should
be used in design with a strong measure of conservative interpretation.
The design of conveyors relies on engineering data on material properties and
material handling interactions. The tables that follow present such information for samples
of (1) raw, municipal solid waste (MSW), (2) shredded, air-classified RDF, (3) a

Table 4 Measured Noise Levels for Refuse Processing

Systems
Location
Tipping floor
Shredder in-feed
Primary shredder
Magnetic separator
Secondary shredder
Air classifier fan
Shop
Control room
Offices
Maintenance laborera
Shredder operatora
U.S. OSHAb (8-hr standard)
U.S. OSHAb (4-hr standard)

Noise levels (dBA)

8592
8590
9698
9096
9195
95120
78
70
67
89
83
90
95

Average worker exposure.

Occupational Safety & Health Standards.
Source: From (462).
b

densified RDF (d-RDF) manufactured in a 13-mm diameter by 10-mm mean length

compressed but friable pellet, (4) the heavy fraction produced in air classification, (5)
the ferrous fraction produced by magnetic separation, and (6) a 1 : 1 (by volume)
mixture of d-RDF and coal. The data were collected in a comprehensive study of the
engineering aspects of conveying (287).
Table 5 indicates the complete list of properties and characteristics that affect
conveyability. Table 6 presents a compact descriptor code referencing several key
characteristics and properties (287). The confidence with which one can extend the
conclusions and recommendations that follow to other refuse types with higher or lower
moisture content, with a different content of light commercial or industrial waste, or with
other critical variations is uncertain. However, the data provide a starting point for
consideration of the materials handling parameters of mixed MSW and its derivatives.
A key measure is the angle of maximum inclination (Table 7) that measures (for a
belt conveyor) the angle relative to the horizontal at which an empty conveyor belt will
successfully transport the material fed to it. If the belt is inclined at or above this angle, the
material being fed will roll back on the belt. Forward motion of material becomes
impossible.
The angle of maximum inclination differs from the angle of repose (Table 8), which
measures the angle between the horizontal and a sloping line from the top of the pile to the
base. It was noted (287) that waste materials seldom form conical piles, so several angles
of repose appear in a given pile. For d-RDF, the angle varied as the pellets broke down.
Loose, degraded pellets produced higher angles than hard, stable pellets. The angle of
surcharge is similar to the angle of repose: the angle to the horizontal the surface of the
material assumes while at rest on a moving conveyor belt. The results in Table 8 reflect
several experimental difficulties. Specifically, the belt was at rest during the measurements

Table 5 Properties and Characteristics

of Solid Wastes Affecting Conveyability
Properties (measured)
Abrasivenessa
Angle of external frictiona
Angle of internal frictiona
Angle of maximum inclination (of a belt)
Angle of repose
Angle of slide
Angle of surcharge
Bulk density, loose
Bulk density, vibrated
Cohesivenessa
Flowability, flow functiona
Lumps: size, weight
Moisture content
Particle hardnessa
Screen analysis and particle size consist
Sized and unsized material
Characteristics (assessed)
Builds up and hardens
Corrosive
Degradable, size breakdown
Deteriorates in storage, decomposition
Dusty
Explosiveness
Flammability
Harmful dust, toxic gas or fumes
Interlocks, mats and agglomerates
Packs under pressure
Particle shape
Stickiness, adhesion
Contaminable
Very light and fluffy, may be windswept
a

Test methods for processed solid waste fractions yet to be developed.

Source: From (279).

and, for economy, both the quantity of waste handled and the dimensions of the test belt
conveyors were relatively small.
The angle of slide (Table 8) is that angle to the horizontal of an inclined flat surface
on which an amount of material will begin to slide downward due to its own weight: an
important measurement in design of chutes or diverters. The data in the table show the
effects of the type and condition of the underlying surface. The state of compaction and
rate of change of tilt in the measurement process also affect the results.
The loose bulk density and the bulk density after vibration for consolidation (Table
9) are other important properties.

Table 6 Material Class Descriptor Codes

CEMA Codea

Material Characteristics
Size

Flowability

Abrasiveness

Characteristics
(assessed)

Characteristics
interlocks

Very fineless than 100 mesh

Fine3 mm or less
Granular75 mm or less
Lumpycontaining lumps 400 mm or less
Irregularstringy, interlocking
Very free flowingangle of respose <19
Free flowingangle of repose 20 to 29
Average flowingangle of repose 30 to 39
Sluggishangle of repose 40 and over
Nonabrasive
Abrasive
Very abrasive
Very sharpcuts or gouges belt conveyors
Builds up and hardens
Deteriorates in storage
Corrosive
Degradablesize breakdown
Dusty
Explosiveness
Flammability
Harmful dust, toxic gas or fumes
Mats down or agglomerates
Packs under pressure
Stickinessadhesion
Very light, fluffy

Conveying Equipment Manufacturers Association.

Table 7 Angle of Maximum Inclination

Solid waste fraction

MSW
RDF
d-RDF
Heavy fraction
Ferrous fraction
d-RDF=coal (1 : 1 by volume)

Flow rate
(ton=hr)

Angle of
maximum
inclination

0.9
0.9
4.5
4.5
4.5
9.1

19
21
30
28
28
27

Note: Belt width, 45.7 cm, belt idlers, 35 degrees, belt speed, 0.51 m=sec.
Source: From (287).

A100
B6
D3
D16
E
1
2
3
4
5
6
7
8
F
H
T
Q
L
N
J
R
V
X
O
Y

Table 8 Materials Handling Properties

Angle of
repose
Solid waste
fraction
MSW
RDF
d-RDF
Heavy fraction
Ferrous fraction
d-RDF=coal
(1 : 1 by volume)

Angle of
surcharge

Angle of slide

Range

Ave.

Steel
plate

Conveyor
belting

20
idler

35
idler

CEMA Codea

25 ' 52
29 ' 49
2746
3059
N=A
4050

39
40
38
40
N=A
42

29.3
31.0
32.8
27.5
N=A
22.0

30.0
35.0
34.5
28.5
N=A
24.0

55
51
N=A
48
N=A
N=A

54
65
49
59
52
40*

E36HJVO
E35HJLXY
D3 35HJQL
E47HQVO
D16 6
D3 45HJQL

*This data point is of uncertain accuracy.

All angles in degrees.
a
Conveying Equipment Manufacturers Association.
Source: From (287)

Table 9 Bulk Density kg=m3

Loose
Fraction
MSW
RDF
d-RDF
Heavy fraction
Ferrous fraction
d-RDF=coal
(1 : 1 by volume)

Maximum

Range

Average

Range

Average

61152
3450
361387
366598
N=A
712

66
43
374
482
N=A
712

66200
3772
402486
334451
194
590

134
54
445
435
194
590

Source: From (287)

1.

Belt Conveyors

Belt conveyors are used for residue in smaller mass burn incinerators and are used
extensively for waste handling in RDF preparation systems. Although this type of
conveyor appears to be simple and its economy prompts frequent selection, there are
many inappropriate and problematic applications in refuse service. In some cases, the
problems are reflections of the irregular or problematic properties of refuse. In other cases,
however, the problem reflects a lack of understanding of the basic principles and working
relationships between the key design variables of the conveyor itself.
SpillageThe loss of material from the belt at transfer points and losses distributed
along the belt path is an important failure mode. For a given material and at
constant mass flow rate, the fraction of the flow that is spilled per unit length of
belt increases rapidly when the belt speed decreases below a critical level. This is
due to the progressive increase in the height of material on the belt as velocity is
reduced. Ultimately, the load begins to overflow. As belt speed increases,

however, spillage also increases due to aerodynamic effects and the vibration and
shock as the load passes the idlers. In general, it is preferred to operate a belt at
higher velocity with a minimum load than the reverse.
In the test program (287), it was observed that belts transporting solid waste
fractions, unless at very low throughput, always generated spillage. Also, it
was noted that 35 idlers were superior to 20 idlers in obtaining increased
carrying capacity for the same velocity and spillage rate. Further, a definite
increase in spillage was observed as the incline of the belt increased. This was
especially the case for the lighter (e.g., fluff RDF) fractions where rollback and
slippage precipitate increased spillage.
TrajectoryKnowledge of the trajectory of material discharging from a belt is
important in the design of discharge chutes, wear plates, and splitters. Studies
(288) of discharge trajectory compared field measurements with a calculation
methodology available in the United States from the Conveying Equipment
Manufacturers Association (CEMA). Agreement was quite satisfactory for RDF
(fluff), MSW, and both heavy and ferrous fractions over a belt speed range
from 0.56 to 2.24 m=sec.
Dust generationData generated in the course of studies of dust generation were
inconclusive. However, a pattern of increasing dust concentration as the conveyor
speed and inclination increased was generally followed. Although the data were
irregular, it appeared that dust control is necessary at transfer points for RDF, dRDF, and RDF-coal blends.
2.

Vibrating Conveyors

Vibrating conveyors are in common use for residue handling in incinerator plants and find
extensive use as in-feed or out-feed conveyors in RDF facilities.
In a vibrating conveyor, a eccentric drive device imparts an acceleration with both
vertical and horizontal components to the conveying pan. Particles on the pan are lifted
upward and pushed forward. Design and operating variables include the frequency and
amplitude (stroke) of the drive. For these conveyors, conveying speed increases as the
frequency and stroke increase.
Since these devices are constructed entirely of metal, they are not damaged by
impact of hot or abrasive materials discharging from a shredder or incinerator grate. Their
inherent operating characteristics lead them to level and distribute the material being
conveyed, which can be beneficial in smoothing out the discharge rate to the next process
step. One should note, however, the consistent tendency among solid waste fractions to
compact due to the vibration of the pan, with MSW and RDF fractions showing the highest
degree of compaction.
3.

Apron Conveyors

Apron conveyors are most often used in resource recovery plants as receiving conveyors
for unprocessed waste or at the discharge of size-reduction equipment where loading or
ballistic impact of material on the conveyor is of concern. The conveyor is made up of a
series of hinged steel pans on a chain and roller assembly. The pans ride on tracks
underneath or outboard of the pans.
The static angle of surcharge on belts is the same as the angle of repose (Table 8) for
steel surfaces. As with a belt conveyor, the maximum carrying capacity is linearly related
to the apron velocity on the flat. Capacity declines with increasing conveyor inclination,

especially at angles above 15 . Spillage is quite limited until angles of inclination are used
where rollback ensues (typically above 30 ).
E.

Size Classification and Screening

Several devices are used to remove fines and to concentrate combustible and noncombustible fractions. These include air classifiers, disk screens, and trommels.
1.

Air Classification

Air classifiers use the influence of a high-velocity air flow to effect separation by
exploiting differences between the aerodynamic and inertial properties of the constituents
of solid waste. The efficiency with which an object subjected to the force applied by the air
flow responds with a directed movement is affected by interference from other objects in
the air stream. Thus, the sharpness of separation is inversely related to the loading rate of
the classifier. The basic separation mechanism of the air classifier is not based on
properties related directly to composition. Thus, the air classifier products streams are
not necessarily greatly enriched in inorganic or organic content. For example, an air
classifier cannot tell the difference between a discarded hammer and a peach or between a
piece of paper and small glass shards. Particles with the same ratio of aerodynamic drag to
inertia tend to end up in the same outlet stream. Nonetheless, air classifiers are very
commonly used in RDF processing as a low-cost first step to achieve a measure of
separation of lights (targeted on paper, textiles, and plastics) from heavies (targeted
on stones, wet materials, glass, and massive metal).
Air classifiers are designed with solids flow predominantly in the horizontal
direction, in the vertical direction, inclined, and in a vertical zig-zag pattern. In all
cases, the design details of practical and effective systems have evolved through trial and
error. However, some general principles have emerged from comprehensive parametric
studies of separation. Table 10 reports the results of studies by the U.S. Bureau of Mines
on the lifting effect of gas flows on several waste categories in a vertical classification
column (411). However, after over 25 years of analytical studies and testing in both
laboratory and field equipment, the ability to confidently predict performance or to design
hardware offering sharp separation is yet elusive (412).
2.

Screens

Screens separate mixed materials into classes based on the ability of particles to pass
through or to be retained on an aperture of a given size. Screens are used alone or in sets to
Table 10 Separation Groupings in Air Classifiers
Velocity
(m=min)
90120
215245
275305
335400
400425

Materials separated
Light fines, dust balls, film plastics
Newspapers and magazines, paper towels and tissue, film plastic, more groundwood
than chemically pulped paper, styrofoam
Some corrugated, more chemically-pulped than groundwood paper
Mostly corrugated
Kraft and corrugated, aluminum, tire fragments, glass, tin cans

Source: From (411)

produce either a simple passdont pass split or to produce an overs stream of too large
particles, an unders stream of too small particles, and a product stream containing
particles in a defined mid-range.
The major types of screens are flat (vibrating, shaking, inclined, or air assisted), disk
screens, and rotary screens (trommels). The last category is described below. The primary
function of screens in RDF processing is to separate oversize for return to size reduction
equipment and to remove undersize that, often, is a concentrate of noncombustible glass,
metal, and mineral matter.
Flat-deck screens are very useful for granular material (sand, ores, etc.) but are prone
to blinding (plugging) with platelike or flaky materials (such as RDF paper fragments) and
to snagging when wires are present. For these reasons, they have found limited use in RDF
applications.
Disk screens are comprised of a series of disks of alternating lesser and greater
diameter that are attached to rotating shaft assemblies arranged perpendicular to the flow
of material. The fines are those particles with a characteristic dimension smaller than the
spacing between the disks. The rotary motion of the disks moves the material along
through the screening cavity. These screens are often used to separate oversize for
secondary shredding (414) and for removal of fines from RDF and from incinerator ash.
3.

Trommels

A trommel is a rotating drum perforated along part or all of its length with circular holes or
slots. Trommels are common and effective elements of many RDF processing schemes. In
this service, they are useful to act as a bag breaker, as a means to break most (not all) of the
glass bottles, and to remove sand, glass, and other fines from shredded RDF streams. The
basic screen construction parameters include the diameter, total length, screening surface
length, inclination angle, aperture size and shape, and open area fraction. The operating
variables are mass feed rate and rotational speed. In a few installations, provision for
variation in the inclination angle is possible. The range of these parameters in trommels
used in RDF applications is illustrated in Table 11.
Analytical studies and tests (376) using raw refuse and the air-classified, light
fraction of MSW were conducted. Comparisons between predicted and experimental data
showed general correspondence between model and test results, but more work was clearly
needed in the analytical development. It was shown that the screening efficiency increases
with decreasing inclination angle (over the range from 2 to 8 ). Thus, using a smaller
inclination angle would be expected to shorten the length of the trommel for equivalent
screening efficiency.
It was indicated that screening efficiency increases with increasing rotational speed.
Also, the rate of change of screening efficiency with respect to rotational speed decreases
with increasing rotational speed within the range of 29% to 73% of the critical rotational
speed (where the material is held on the wall by inertial or centrifugal forces). The critical
speed Nc for diameter (D) in meters is given by the dimensional formula
Nc

251:1
D

rpm

This suggests that, to a degree, poor screening efficiency due to overloading or insufficient
length of screen can be compensated for by increasing the rotational speed.

Table 11 Dimensions and Capacities of Trommels

No.

Diameter (m)$
length (m)

1
2
3
4
5
6
7
8
9
10
11

3:35 $ 12:19
3:66 $ 8:53
3:66 $ 11:58
3:05 $ 15:54
3:05 $ 13:72
3:05 $ 7:92
3:05 $ 6:40
2:74 $ 9:14
2:44 $ 6:10
2:74 $ 9:14
2:44 $ 13:41

12
13
14
15
16
17
18
19
20
21
22
23
24

2:44 $ 10:67
2:13 $ 6:10
2:13 $ 6:10
2:13 $ 4:54
1:83 $ 7:32
1:52 $ 6:10
1:22 $ 5:49
1:22 $ 2:44
1:22 $ 3:66
1:22 $ 3:35
1:22 $ 2:44
0:91 $ 4:88
0:91 $ 3:05

Hole size
(mm)

Capacity
(ton=hr)

120, 63, 19, 6

32
34
57
120
40, 210$254
Various
120
16
19
10 $ 51, 25 $ 101,
203 $ 254, 152
70, 26
Various, 13
102, 51
22
19
102, 32, 16
19 or 76
13
132
19
76, 19
10, 13
102, 51

68.2
40.9
90.9
50.1
56.8
9.1
4.5
22.7
22.7
36.4
30
45.5
40.9
45.5
9.1
18.2
4.5
13.6
4.1
6.8
12.3
13.6
9.1
3.6

Power
(hp)

Speed
(rpm)

40
20
120
60
80
6

17
25
20
88

11
9
10.5
10
11
930
020
14
14
11

50
2
10
12
15
15
15
7.5
7.5
7.5
15
2
1.5

13
28
11
25
20
816
22
1030
120
1236
443

Source: From (413)

F.

Ferrous Metal Separation

Ferrous metal separation is practiced both ahead on (in the preparation on an RDF) and
after combustion. In both cases, either permanent or electromagnets are used to effect
separation. Figure 2 shows the relationship between the diameter of the magnet and the
working distance to an object containing ferrous metal. The separation effectiveness of the
magnet depends on the distance between magnet and metal, the weight of the object
relative to the magnetic force, and the depth and weight of any overburden of nonmagnetic
refuse material covering and=or entangled with the magnetic material.
The layout of ferrous metal separations is usually of two types:
1.
2.

A continuous belt fitted with magnets (a belt magnet) operating above a

conveyor handling mixed waste and pulling ferrous metal from the conveyor.
A drum fitted with electromagnets (a drum magnet) operating at the end of a
conveyor. The magnet is energized only for a preset fraction of the drum rotation
such that ferrous metal is held longer than nonferrous material and, thus, is
released with a different trajectory. The ferrous is directed to a special receiving
chute.

Figure 2 Working distance of magnets.

In both cases, some nonferrous is inevitably included in the product stream due to
entanglement, physical trapping, etc. Thus, if a high-quality scrap is desired, a second
separation of the ferrous-enriched product is needed.
II. SLUDGE HANDLING
A.

General

Pumps of most conventional designs can, in some circumstances, be used in (dilute) sludge
pumping service. With wastewater sludges, progressive cavity (for sludge up to about 20%
solids) or piston pumps (for the range of sludge solids content) have shown the best
performance. The special rheological properties of sludge require special analysis,
however, for the reliable estimation of pressure drop (see below).
Conveyance of dewatered sludges via belt, tubular, and screw conveyors, slides and
inclines, and elevators has been demonstrated. Because the consistency of the sludges is so
variable, however, the design selected must consider performance under conditions widely
variant from average.
Belt conveyors using field-vulcanized seams are simple and reliable and, for
semisolid sludges, can operate at up to an 18 incline. Skirtboards are recommended at
critical areas. Adjustable-tension finger-type scrapers mounted beyond the idler on the
flattened portion of the belt are recommended. Splashing and impact at transfer points
should be minimized. In this regard, the thixotropic character of sludge (significant
decreases in apparent viscosity under shear) can produce surprising results when, after a
splashing impact at a transfer point, a stiff sludge cake can turn to watery liquid.

Screw conveyors are useful for sludge conveyance on the horizontal and, depending
on the sludge consistency, up inclines. Abrasion-resistant construction, provision for easy
inspection, and maintenance ingress are recommended. In most cases, internal, intermediate bearings are undesirable, thus limiting the maximum conveyor length to
approximately 20 ft.
B.

Sludge Pumping in Pipes

Many sludges can be pumped; perhaps after heating to reduce the viscosity. However, the
high-pressure drop associated with sludge pumping requires that careful attention be given
to estimation of the flow characteristics.
Unless a sludge has been dewatered to a significant degree, it can be transported
most efficiently and economically by pumping through pipelines. Further, the use of
pumps to feed sludge to incineration systems has become the method of choice for three
commanding reasons:
1.
2.
3.

Feeding through a pump produces a stable feed rate that results in lower air
emissions, reduced fuel use, and more troublefree operation of the incinerator.
Feeding through a pump eliminates several limitations in layout over the use of
conventional screw or belt conveyors.
Feeding through a pump introduces sludge to the incinerator without admission
of any unwanted tramp air and with little or no exposure of the sludge, thus
facilitating the containment of odors.

Head losses must be estimated for sludge pumping, preferably using rheological data for
the specific sludge of interest, since friction pressure drop behavior is often not the same as
for water; especially for sludges of greater than 2% solids content. Head requirements for
elevation and velocity, however, do parallel those for water.
In the well-behaved, Newtonian flow of water, oil, and most single-component,
single phase fluids in laminar flow situations, the pressure drop is directly proportional to
the velocity and the viscosity. Further, the viscosity is a constant, independent of velocity
(velocity being a measure of the shear rate in the fluid). The relationship between the
laminar flow pressure drop DP (atm) per length L (m) of pipe of diameter D (m) for a fluid
of viscosity m (centipoise) flowing at a velocity V (m=sec) is given by Poiseuilles law:
DP

3:157 $ 10'7 mVL

D2

As the velocity increases in a given flow situation, it is found that the flow behavior departs
from purely laminar characteristics through a transition region where eddy formation
increases in frequency and severity until a turbulent condition is attained. The onset of
eddy formation is associated with the attainment of a Reynolds number (NRe and
dimensionless) of about 2000, and fully turbulent flow is observed at Reynolds numbers
above 4000. The Reynolds number is calculated as
NRe

r0 u0 d0 d0 G

m0
m0

where the fluid density r0, the mean velocity u0, the characteristic dimension of the flow
(typically the diameter d0 for pipe flow), the fluid viscosity m0, and=or the mass flow per
square meter G are in consistent units. The Reynolds number is one of several

dimensionless groups of process variables used in the characterization of fluid flow and
other processes. Analysis (and, confirming experiment) often shows that these groups are
physically meaningful parameters such that similitude between two situations is found if
the numerical value of the group is maintained identical. The Reynolds number can be
viewed as the ratio of momentum forces in the flow field (characterizing turbulent eddy
processes) and viscous forces (characterizing laminar processes). Thus, Reynolds number
similitude is aimed at maintaining a similar balance between eddy and laminar flow
behavior.
In turbulent flow, the pressure drop DP (atm) per length L (m) of pipe of diameter d0
(m) for a fluid of density r (kg=m3 ) and viscosity m (centipoise) at a velocity V (m=sec) is
given by:

DP

fV 2 Lr
35:46gc d0

10

where gc is a conversion factor equal to 98.06 kg-mass meters per kg-force sec2 . The
variable f , the Fanning friction factor, is a function of the Reynolds number and pipe
roughness.
Different behavior is observed for the flow of sludge. The basic fluid behavior is
non-Newtonian for wastewater sludges, Specifically, viscosity is not constant, as
determined by tests in a rotational viscometer. In this type of machine, the test liquid is
placed between concentric cylinders, one of which rotates. The torque on the non-rotating
cylinder is measured as a function of rotational speed. In such a device, sludge shows a
reduction in apparent viscosity under shear (thixotropic characteristics).
When testing sludges, it is usually found that there is a yield stress t0 below which
the cylinder will not start to rotate (see Fig. 3). Then, as the rotational speed (shear rate) is
increased, the resistance at first increases and then decreases. Such a reduction in apparent
viscosity with shear rate is typical of thixotropy and results from the breakdown of
physical structures and interparticle attractive forces. Once these structures are broken
down, one can characterize the fluid property akin to and having the same units as
viscosity as Z, the coefficient of rigidity. This flow behavior is known as that of a
Bingham plastic. For such materials at high shear stress, the flow behavior compares to
that of a Newtonian fluid, as shown in Fig. 4.
To describe the pressure drop in such systems, one must calculate the starting
pressure needed to overcome the yield stress and the pressure to overcome friction. The
starting pressure is given by

DP

4Lt0
d0

11

It should be noted that, often, the yield stress (t0 ) increases with time as the material rests
in the pipe in the no-flow situation. Therefore, startup pressure requirements may be
considerably greater than those calculated based on the t0 developed as shown in Fig. 3.
Since the development of such high-yield stresses is time-dependent, consideration should
be given to purging the line (especially pump suction lines) if extended periods of no flow
are encountered.

Figure 3 Viscometer test of sewage sludge. (From [223]).

To calculate the pressure drop for steady flow, two dimensionless numbers are used:
a modified Reynolds number given by
0
NRe

r0 u0 d0 d0 G

Z0
Z0

12

and the Hedstrom number NHe (the product of the Reynolds number and the ratio of the
yield stress to the viscous force) as given by
NHe

d02 t0 gc r0
Z20

13

Equation (9) can be used to estimate the pressure drop DP in atm for r0 in kg=m3 , the
length L in m, the velocity V in m=sec, and using Z0 in centipoise as the Bingham plastic
limiting value (at high shear rate) for the coefficient of rigidity. The overall Fanning
friction factor ( f ) in Eq. (9) should be developed (207) as a function of the friction factor
for laminar ( fL ) and turbulent ( fT ) flow by:
f fLb fTb 1=b

14

where fL is the (iterative) solution to Eq. (15):

4
7
fL 16=NRe0 (1 1=6NHe =NRe0 ' 1=3NHe
=f 3 NRe
0 (

15

Figure 4 Comparison of behaviors of wastewater sludge and water flowing in circular pipelines.
(From [223]).
with
'0:193
fT 10a NRe
0

14

and where
a '1:471:0 0:146 exp'2:9 $ 10'5 NHe (

15

b 1:7 40;000=NRe0

16

and
An estimate of the friction factor can also be taken from Fig. 5.

Figure 5 Friction factor for sludge analyzed as a Bingham plastic. (From [223]).

8
Incineration Systems for Municipal
Solid Wastes

If a gas- or oil-fired boiler is compared with boilers for the combustion of coal, one
observes substantial differences beginning with fuel receipt, handling, and storage through
to the stack. Clearly, departing from a pumpable, atomizable fluid to a solid and from a
low-ash to a high-ash fuel introduces pervasive and costly changes to the design,
construction, and operating procedures. A similar comparison in the area of waste
incineration yields more dramatic results.
Even more profound differences between systems for gases or liquids and those for
solid wastes are seen if one examines the individual unit processes and unit operations that
take place from the time a waste enters the disposal facility to the time its combustion
products leave. Generally, it is found that more steps and steps of greater complexity occur
for the solids. In kind, the design problem for the sludge or solid waste incineration
systems is found to be more empirical, less tractable to confident prediction, and more
subject to partial or total failure than their counterparts for gaseous or liquid wastes.
Further, in many cases key solid waste parameters are subject to unpredictable short- and
long-term variation. Municipal solid waste is heterogeneous, irregular in key materials
handling properties, highly and unpredictably changeable, and always problematic. One
facility engineer expressed his frustration with the challenges of waste handling with the
caution, If the waste can possibly get you, it will! That word of warning should be
particularly heeded in the design of materials handling equipment.
The following discussions are intended to describe contemporary incineration
practice for municipal solid waste combustors (MWC). In a few instances, particularly
relating to flue gas temperature control for refractory systems, some additional process
analysis methodology is presented.
The predominant concern here is on incinerators for municipal solid waste. Because
of the wide variation in waste character in industrial situations, systems for industrial
wastes often require designs where materials handling, construction materials selection,
effluent gas quality goals, and other features must be tailored to the unique nature of the
industrial disposal problem. The complexity of the resultant designs and the many new
dimensions in combustor configuration are generally beyond the scope of this volume.

The objective of this section is to answer the difficult question: What is a municipal
incinerator? In the early years of refuse incineration in the United States, incinerators were
designed in a technical collaboration between the public works department of the owner
city (county, etc.), its consulting engineer, and the major component vendors (especially
the grate and=or boiler manufacturer). The 1970s saw the emergence of a new entity: the
system vendor. Unlike the earlier situation, the governmental entity with a waste disposal
requirement now found itself in a commercial environment where single-point, overall
responsibility for the design and, often, the operation of the facility could be placed in the
hands of one of several competing firms. The final working relationship was codified and
detailed in a comprehensive contract document (the Service Agreement). The role of the
consulting engineer had largely shifted to project planning, assistance in financing,
preparation of permitting, and preparation of performance specifications for competitive
bidding by several of the system vendors. Elsewhere, entrepreneurial system vendors took
the lead in developing a project.
The Service Agreement often goes beyond a simple documentation of a contract to
provide waste incineration services. Since the lifetime of the agreement is often 15 to 20
years, many of the circumstances defining the nature of the service, economic factors,
environmental requirements, and other important parameters will change. Thus, the
agreement defines the set of reference system characterizations that were the basis of
the original procurement and indicates methods and guidelines with which to update the
cost and=or performance basis from the baseline. The characterizations include a
reference waste composition and heat content; unit costs for labor, utilities, taxes, and
reagents; environmental requirements; and daily and annual processing rates and energy
recovery targets.
The new, system vendor-dominated incineration business employs a wide variety of
designs to do the same job. This individuality reflects both the growth of incineration
technology in recent years and the large number of basic design parameters that are
somewhat flexible and can be bent to the prejudices of the design firm. The systems used
can be divided into two broad categories: mass burn technology burning raw,
substantially unprocessed refuse and refuse-derived fuel (RDF) technology, where a
prepared, refuse-based fuel is burned. Although mass burn technology dominates the
market in the United States and Europe, both approaches have their strong points and their
advocates.
The discussion that follows outlines the options in mass burn incinerator design and
the special characteristics of RDF-based combustion systems. The RDF preparation
technology is, substantially, left to other books although refuse processing is briefly
covered in Chapter 4. Any one of the pages or even paragraphs would require expansion to
a chapter or book in its own right if the subject were dealt with at a level of detail fully
supporting hardware design. It is hoped, however, that after reading this section the reader
will have familiarity with many of the terms involved in describing an incinerator and with
the wide variability possible between systems.

I.

PERFORMANCE OBJECTIVES

The performance objectives of a municipal waste incineration system are

To process each normal operating day not less than the quantity of waste with an
analysis and heat content specified in the Service Agreement

To process the minimum weekly, monthly, and=or yearly quantity of waste specified
in the Service Agreement
To consistently operate within the emission limits and other legal constraints of all
applicable environmental regulations to include restrictions on the concentrations
or mass rates of air or water pollutants, sound pressure levels, and=or the
maintenance of specified system operating parameters within designated limits
To protect the health and well-being of incinerator employees and of the commercial
and residential community that abuts the operation
To protect the capital investment reflected in the equipment, buildings, roads, etc.,
comprising the incineration facility such that the useful operating life and
maintenance and operating expenses of the incinerator are not adversely impacted
To meet any production guarantees regarding residue quality and quantity; export
rates of power, steam, or other energy-related products; or other commercial
promises
The achievement of these objectives is strongly supportive of a healthy plant operation,
good customer relations, and good financial performance.

A.

Throughput and Refuse Heat Content

Many of these performance objectives are strongly influenced by the characteristics of the
waste. The most basic connection is through the heat content of the waste material since, in
essence, an incinerator is a system to process heat. Therefore, the capacity of an incinerator
is intrinsically associated with a maximum heat release rate (the maximum continuous
rating, or MCR) and not a mass throughput rate (except as the mass rate, multiplied by the
waste heat content, is equivalent to a heat release rate). Unfortunately, many municipal
clients believe that their contract relationship with the incinerator operator is a commitment
to process a given mass of material (e.g., 400 tons per day) rather than to process a
specified number of millions of kilocalories per day. Experience has shown that if this
potential misperception is not clearly addressed in the Service Agreement, changes in waste
heat content over the contract life will lead to customer dissatisfaction and, even, lawsuits.
Why is heat release the real variable? Figure 1 illustrates the process and hardware
connections that spring from the MCR heat release parameter. The heat release rate,
because of the approximate equivalence between heat release and combustion air quantity,
is strongly related to the volumetric flow rate of combustion air and of the products of
combustion. Thus, the heat release rate sets the size of the forced draft and induced draft
fans, sets the size of the air pollution control system, and sizes the ductwork and dampers,
pressure drops, etc., throughout the flow system. Also, the heat release rate, for a given
combustion chamber, strongly impacts the heat transfer rates (both convective and
radiative), which affects the temperature of surfaces in boilers and on refractory walls.
Thus, exceeding the design heat release rate can result in overheating of critical system
components. All these factors illustrate why incinerator capacity is quite properly equated
to the MCR rather than the tons fed.
Another limitation on capacity relates to the structural strength and materials
handling capabilities of the grate and the physical dimensions and capacity of the residue
and fly-ash handling systems. These physical limitations generally limit the furnace
throughput to approximately 110% of the basic design capacity.

Figure 1 Maximum continuous rating (MCR).

B.

The Firing Diagram: The Overall Process Envelope

The firing diagram shown in Fig. 2 provides a concise, graphical statement of the operating
process envelope of an incineration system. Specifically, the area bounded by the dotted
lines represents the combinations of mass feed rate and refuse heat content supported by
the referenced incineration furnace. For all points within the dotted area, the furnace can
meet its design mass disposal rate and still remain at a technically sound fraction of MCR
and the physical throughput limitations.
Let us consider the various elements of the boundary of the operating zone:
Maximum heat releaseThe horizontal top line of the zone is the MCR. Heat
release rates over this limit unduly stress the equipment or exceed design limits
for fans, air pollution control equipment, etc. Also, in waterwall boiler systems,
operation above this heat release rate may lead to boiler-tube failures, tube
erosion, etc., contributing to unscheduled outage.
50% of MCRThe horizontal bottom line of the zone is set at 50% of the MCR.
While somewhat arbitrary, burning at less than one-half the design heat release is
often accompanied by poor mixing (increasing CO and hydrocarbon pollutant
emissions), degradation in residue quality, furnace control problems, draft control
problems, etc.
110% of maximum throughputThe vertical rightmost boundary of the zone is
set at 110% of the design mass throughput. This is a reasonable estimate of the
maximum feed rate that can be accommodated by the grate and residue handling
system.
50% of maximum throughputThe vertical, leftmost boundary of the zone is set
at 50% of the design mass throughput, reflecting the constraint that as the
throughput drops from the design level, it becomes more likely that the grate will
be exposed to furnace radiation. Also, the breakdown in the performance of the
solids materials handling equipment becomes more likely.
Maximum rate of highest heat content refuseThe sloped top boundary of the
zone is set by the heat releasethroughput line for the highest heat content refuse.
This line intersects the MCR line at the design throughput line. This is the
maximum heat content refuse used as the basis of design in setting the MCR.
Note that for this heat content refuse, the system can just meet the design
throughput rate (often equal to the minimum rate set in the service agreement)
and stay within the MCR.
Maximum rate of lowest heat content refuseThe sloped bottom boundary of the
zone is set by the heat releasethroughput line for the lowest heat content refuse
that intersects the 50% MCR line and extends to the design capacity limit.
Since the operating zone described in the firing diagram is a simple, graphical statement of
the maximum operational capabilities of the incineration system, there are merits to
including the diagram as part of the Service Agreement.
C.

Plant Availability

Plant availability is a critical characteristic of an incineration facility. Reliable waste

disposal is, clearly, important to public health and community sanitation. From the
standpoint of the facility contract operator, processing waste is, generally, the only way
the facility generates revenue. If the facility is off-line, cash inflow stops while fixed costs,

Figure 2 400 TPD incinerator firing diagram.

labor, taxes, and other expenses continue. Thus, the design process must give careful
consideration to the tradeoffs between increased capital investment operating expense and
improvements in on-line availability.
1.

Power Outages

Although the effects of reduced plant throughput on annual throughput guarantees due to
electrical power outages may fall within the force majeure clauses of the service agreement
(referring to acts of God for which no blame is assessed), the facility still suffers from
the cessation of the revenue stream. If the historical pattern of outages in a proposed
electrical service area suggests that outages may be frequent and of extended duration,
there may be ready justification for some measure of protection. This could include:
Equipping major power users (e.g., forced and induced draft fans, boiler feedwater
pumps, stokers) with both electrical and steam turbine drives. The electrical
system would be necessary for initial black startup. Thereafter, the systems
could be switched to steam drive for sustained operation. Although this approach
increases capital investment, it insulates the plant from the impact of power
outages and reduces the cost for purchased electricity.
Installing on-site diesel power generation (for the computer, install a battery pack as
a noninterruptible power supply) for minor electrical users (small motors, critical
lighting, etc.) as well as for the furnace grates and cranes. Alternatively, one could
increase the standby power capacity to a size sufficient to operate the entire plant.
In the design process, the degree to which the facility is provided protection against power
outages should be drawn from an engineering tradeoff analysis. Such an analysis looks at
life cycle costs with installed power generation equipment (including consideration of the
avoided cost for electricity that did not have to be purchased) in comparison to the lost
revenue for disposal fees and energy.
2.

Equipment Outages

A comprehensive analysis of the mechanical reliability of all proposed structural and

process components can serve to pinpoint the subsystems that could potentially shut down
an incinerator line or the entire plant. These could include, for example, outages of the
crane and grapple, refractory failure, feedwater pump failures, and so forth. If possible,
maintenance data summarizing the frequency and severity of outage experience in similar
plants should be reviewed. Solutions to such problems can include several strategies:
Installed backups: Many plants install parallel components if outages are frequent or
prolonged and, especially, if the overall plant availability is impacted. Examples
of such components include the bridge crane, cables and grapple, and boiler
feedwater pumps. For the air pollution control (APC) system, redundancy (e.g.,
spare modules in the fabric filter) may be appropriate even though the capital cost
is high since in most jurisdictions, APC outages trigger immediate shutdown.
Spare parts: If the turn-around time to install a new item is short, consideration can
be given to maintaining a larger inventory of spare parts to facilitate rapid return
to operations. This option is usually more appropriate for components of a single
line (e.g., a combustion air fan motor) than for assemblies that impact the overall
plant. An inventory of replacement refractory bricks and castable material is
clearly necessary. Spare grate bars are similarly appropriate.

Response strategy: The importance of plant availability warrants detailed planning

for all significant outage scenarios. Plant operators and maintenance staff should
receive training in the most effective and rapid response to minimize collateral
damages and to reduce the time to return to service. Such planning along with
regular inventory and restocking of spare parts and maintenance supplies and
tools will keep the staff both ready and capable to respond to problems.
Preventive maintenance: The working environment in an incineration facility is
aggressive and unrelenting. Consequently, the availability of individual items of
equipment and of the plant is closely tied to the commitment of the facility
management and staff to an aggressive and effective program of preventive
maintenance. This means expenditure planning, staff scheduling and training,
strict maintenance scheduling, and appropriate prioritizing of the maintenance
function.
In the design process, the degree to which the facility provides protection against
equipment outages is drawn from an engineering tradeoff analysis. The analysis should
look at the life cycle costs with backups, spares, etc., and include consideration of lost
revenue for disposal fees and energy.

II. SITE DESIGN CONSIDERATIONS

The design of the incinerator site significantly impacts the cost of the facility, the control of
storm water runoff, and the efficiency of the truck traffic flow. Site development costs
typically represent 2% to 4% of the total cost of a plant, but can increase if substantial
earth moving is required for the site. Storm water must be controlled to limit the impacts of
plant runoff on nearby receiving waters (streams, rivers, and lakes). Traffic backups
(queuing) can occur if the plant roadway system is not configured properly. The designer
must consider all of these site issues when evaluating different sites for the plant and in
orienting the plant on the selected site.
A.

Site Grading

The incinerator structures should be located to minimize the need for extensive site
grading. Grading is expensive and requires substantial time to complete. The waste
receiving area (tipping floor or pit charging level) of the plant is generally located at a
higher elevation than the combustion system to allow the waste to be fed into the
incinerator by gravity. Thus, the plant is usually oriented so that the waste receiving area is
located on the highest elevation of the site, and the combustion and air pollution control
equipment are located toward the lower elevations.
The roadway grades should be limited to allow loaded waste delivery trucks to easily
start and stop when the waste delivery area is backed up. Typically, roadway grade is
limited to 5% or even less if many uncovered, open-top delivery trucks are expected at the
plant and spillage could occur.
The grade of planted areas should be limited to minimize soil erosion during heavy
rains. Typically, planted area grading should be limited to a slope of 4 : 1. Steeper slopes
may be employed, but special measures for slope stability and erosion control may be
necessary.

B.

Site Drainage

The drainage system design should control storm water runoff from the plant site. Since
uncontrolled storm water can carry pollutants from the ash loading, chemical handling,
truck washdown, and waste receiving areas into nearby receiving waters, containment and
diversion to sewers or on-site treatment or holding facilities are appropriate. Typically, the
control of storm water runoff is most critical at the ash handling areas. Many plants have
installed first flush collection and sedimentation tanks and sand, oil, water separators to
collect the ash from the storm water runoff.
Chemical handling and spill containment areas must control the runoff of storm
water so that spilled or residual chemicals cannot be washed off-site. To restrict the flow of
rainwater if chemicals are present in the containment area, valves are often placed on the
spill containment area drains. Holding ponds can be used to temporarily store large
quantities of storm water runoff if the off-site drainage systems do not have sufficient
capacity
C.

Site Traffic and Road Considerations

The plant roadway system should minimize traffic congestion and allow adequate room for
queuing of trucks at the scale house and at the tip area. The plant traffic should all flow in
one direction around the plant. Roadways should allow the waste delivery trucks to enter
and exit without stopping or crossing the plant roadways. Roadways used by waste
delivery and ash haul trucks should have a minimum turning radius sized for the largest
expected vehicle in use and should be wide enough for two trucks to pass. Truck
congestion is common at the scale house and at the entrance to the tipping area, and
plant roadways should provide space for at least 10 trucks to line up at both locations.
III.

COLLECTION AND DELIVERY OF REFUSE

In the incinerator system, the handling of refuse begins with the delivery of materials to the
site. In the United States, most refuse is delivered to the site in motor vehicles although
railroads, private automobiles, or barges are sometimes involved. Delivery vehicles can
include open dump trucks, commercial vehicles, and private cars. By far, however, most of
the waste delivered to incineration sites is conveyed using specialized trucks with
equipment for compression and densification. In some instances, very large (50- to
70-m3 compaction truck trailers are employed to ferry refuse from centrally located
transfer stations to a distant incinerator site. Such transfer stations are serviced by smaller
collection vehicles.
The refuse loaded into the original collection vehicles usually has a density of about
80 to 240 kg=m3 , depending on the nature of the refuse and how it has been packed at the
refuse generation point. Refuse can be loosely placed in collection receptacles, manually
stamp-packed, or compacted at the site by hydraulic compression and=or baling devices. In
a few cases, the waste is ground or shredded. A power compaction unit can compress the
refuse at the generation site (such as at commercial establishments, hospitals, apartments,
hotels, etc.) to a density of about 500 to 1000 kg=m3 . However, most refuse loaded into
vehicles at the collection site is compacted in the truck body to 250 to 500 kg=m3 by
means of mechanical and=or hydraulic systems using power takeoffs from the vehicle
engine. Compaction is employed in most municipal and private collection vehicles to
increase the payload capacity of the vehicle.

Most incinerators have a scale to weigh the incoming refuse. Various types of
scales are available. Some are manually operated with the weights recorded by the scale
operator. Others furnish a printout of the gross, tare, and net weights, together with
the license or code number of the vehicle. Larger modern installations use a computerized scale for automatic weighing: The driver of the incoming vehicle inserts a coded
card to supply the vehicle code number and the tare weight of the vehicle, and the
automatic system prints the gross weight and net weight, as well as the date and time of the
delivery.
The plant should have an accurate and reliable truck scale to weigh and record
all incoming waste deliveries and outgoing ash trucks. Plant personnel and=or local
municipal officials must supervise the scale operations to ensure that all trucks travel over
the scale and are weighed properly. The tare (empty) weight of the waste delivery trucks is
used to calculate the net weight of the waste in each waste delivery load. Weigh scale
systems can be connected to a computer that records the waste delivery data. Such a scale
data recording system can print a ticket to provide the truck driver with a record of the
delivery.
The scale operator is also responsible for determining if the trucks are authorized
to deliver waste to the plant and is the first line of defense in assuring that the type
of waste is compatible with the environmental permits and the technical capabilities of the
system. Clearly, the reliability and integrity of the scale operator are key to a successful
operation.
Typically, a small building (the scale house) is provided adjacent to the scale to
provide shelter for the plant personnel and for the weigh scale computer system. Several
plants house the scale operator and weigh scale computer system in the plant administrative offices when the scale has been located adjacent to the administrative building and
the lines of sight are designed with this function in mind. A typical weigh scale is 10 m
long and 3 m wide and has a gross (total) capacity of 55,000 kg.
The waste loads should be periodically inspected when they are dumped to identify
material that, in accord with the operating permit of the incinerator and the provisions of
the Service Agreement, cannot or should not be processed. Any oversized or hazardous
material should be removed immediately and temporarily stored in a safe area. In some
plants, hand-held radiation detectors are used to identify and remove radioactive material.
Since the charging of material to the incinerator which is forbidden by air permits or other
regulatory limitations places the incinerator operator at risk for fines, penalties, or
shutdown, the importance of a good faith implementation of waste load monitoring is
critical.
The receipt of refuse also presents the key to preventing fires and hazardous events.
Some plants use hand-held combustible gas detectors to identify and remove potentially
explosive items. Such screening is particularly important for plants that process the refuse
to a fuel (RDF) since the shredding stages of fuel preparation often expose the waste to
sparks and other potential ignition sources. Provisions and penalties should be clearly
understood, prohibiting the delivery of all unacceptable wastes. These would include
hazardous materials, explosives, chemical waste, flammable liquids, and, perhaps, medical
and\or pathological wastes. Plant policies (and permit provisions) should be visibly
displayed in signs posted at the point of entry. These passive reminders need to be
reinforced with regular visual inspections by the truck scale operators, random selection of
incoming trucks for dumping and visual inspection of contents, and constant vigilance by
tipping floor personnel and crane operators.

IV.

REFUSE HANDLING AND STORAGE

Refuse is a perverse material with unusual physical and mechanical properties. Unlike
many solids, it does not cone when piled up and often exhibits a negative angle of repose
(an overhang). Because of its source and moisture content, portions of the waste can be
sticky so it may not discharge cleanly from belt or apron conveyors. This can cause
housekeeping problems along the return run. Waste compresses, which can result in
binding at points of close tolerance in moving machinery. Since the compressed material is
not brittle and thus does not break off, the tough, compressed waste can keep gates and
doors from closing fully.
Waste physical forms include fine powders, sometimes dry and dusty and sometimes
wet or even watery; long streamers (e.g., plastic ribbons); individual pieces of paper that
are easily blown about and heavy, thick telephone books; cans, bottles, and other
containers carrying explosive, toxic, corrosive, or flammable liquids; long and tough
shapes such as long steel bars, thick wood planks or beams; and massive materials from
engine blocks to furniture and mattresses. This variability emphasizes the special need for
rugged and flexible materials handling equipment.
A.

Tipping Floor-Based Waste Storage and Reclaim Systems

In small incinerator plants (say, less than 150 tons per day), a paved tipping floor is
sometimes used instead of a receiving pit. Under this concept, the refuse is dumped
directly onto the floor by the collection vehicles and marshaled into piles using front-end
loaders. On demand, the waste is reclaimed from the storage piles and charged to the
incinerator(s).
The tip floor area for the storage piles must include a concrete push-wall so that the
pile does not move when a front-end loader picks up a load. In order to avoid exceptionally
long holding times, the storage area should be large enough to create two separate storage
piles so that new waste deliveries can be stored in one pile as the incinerator is fed from a
pile of older waste. Otherwise, the waste storage pattern is on a first in, last out basis that
encourages fly and other vermin infestation, odor, and even spontaneous combustion.
Typically, the maximum pile height is 3.5 to 4 m depending, importantly, on the maximum
lifting capacity of the front-end loader. Typical waste density is 200 to 250 kg=m3
depending on the moisture content and makeup of the material.
The preferred location of the two storage piles is in the corners of the storage floor
on each side of the incinerators. The area between the incinerators should be left open to
allow the front-end loaders to maneuver and to provide access for fire control. The
maximum reach of the front-end loader and the height of the concrete push-walls
determine the height of the storage piles on the tip floor. The concrete push-walls
should be a minimum of 3.6 m high and designed to withstand a force of 600 kg=m2 .
The floor surface area in front of the waste feed hoppers can be expected to wear
down from the scraping of the front-end loader bucket. Regular reinforced concrete usually
wears down to the steel reinforcing bars in 8 to 10 years. There are options to eliminate the
expensive repairs of a new tipping floor section, including
A concrete hardening additive can be used in front of the hopper to minimize the
wear.
A steel plate can be located above the hydraulic waste loading ram in front of the
loading hopper that can easily replaced. Steel rails can also be cast in the concrete

approximately 1 to 2 cm above the concrete to keep the front-end loader bucket

above the concrete surface.
The floor dump approach is low in initial capital cost in comparison to the pit and crane
design. Also, the floor dump facilitates visual checking of incoming refuse for excluded
wastes (e.g., hazardous materials and automobile batteries). However, fire control is
difficult in the floor dump situation (especially if the fire starts at some distance from
the front face). The floor becomes slippery from the dumping and presents a safety hazard
to employees. Another hazard comes from the high levels of dust swirling about the
workers who, perforce, must work for long periods of time in the dumping areas.
B.

Pit and Crane-Based Waste Storage and Reclaim Systems

In larger incinerator plants, refuse is most often received and stored in a pit below ground
level. A traveling crane with a bucket is used to pile the refuse for storage and to move it
away from the area just below the unloading area so that the pit can accommodate
additional refuse. The bridge travel covers the entire length of the storage pit, and the
trolley travel covers the width of the pit. The bridge crane design criteria must take into
consideration the reliability needs of the incinerator, which are comparable to cranes
designed for steel mill duty (class F): nearly the most stringent in the industry. Industrial or
modified industrial-duty cranes are unacceptable in incinerator service.
The crane and bucket are also used to feed the incinerator furnace. Generally, the pit
is sized to hold the quantity of refuse that can be burned in 2 to 3 days. In most modern
plants, the crane is controlled by one or two crane operators working from a pulpit
mounted on one of the end-walls or in the middle of the back wall of the pit. The duty
cycle of a refuse handling crane should conform to minimum requirements established by
experience, which show a charging time of 30% to 35%, a restocking time of 35% to 50%,
and an idle time of 15% to 35% (461). All bridge and trolley motors should be sized for at
least 115% of the design load, and the hoisting and grapple motors should have a rating of
at least 125% of the design load.
Three types of buckets are in common use for handling municipal refuse: the
clamshell bucket, the grapple bucket, and the orange peel bucket. The clamshell and
grapple buckets can be equipped with teeth to assist in digging into the refuse for a full
bucket. The orange peel bucket does not require teeth to grab the refuse, even for unusual
or large pieces. With the orange peel, it is difficult to thoroughly clean the floor of the
receiving pit. The clamshell and the grapple are well suited to a clean-out of the pit,
providing the teeth are first removed. A steel plate is sometimes used to cover the teeth
when cleaning the floor of the pit with a grapple bucket. Pit cleaning is important for
sanitation, good housekeeping, and elimination of odors from decomposing refuse. It is
prudent to design the pit and crane facility so that undesirable material inadvertently
dumped into the pit can be grasped by the bucket and carried to a point where the reject
material can be easily shunted to a truck or otherwise eliminated.
While open receiving areas are possible, they are infrequently used. The receiving
area should be enclosed to avoid windblown refuse from causing a nuisance, to control
dust and odors, and to prevent snow and ice from interfering with the movement of
vehicles. Enclosing the refuse receiving area also assists in refuse moisture control by
keeping the rain off. Moisture stability is helpful in obtaining uniform combustion
temperatures in the furnaces.

Fires in the pit are not uncommon. Generally, the fires are started by dumping
material that is already smoldering or burning (a hot load"). Spontaneous combustion or
buildup of heat from composing of refuse components is another potential cause. It is
therefore necessary to equip the pit area with fire hoses at several strategic and accessible
locations surrounding the pit. Hoses permit focused application of water to particular hot
spots rather than the more general quenching effects of sprinklers. If circumstances
support this approach, the fire may often be extinguished without undue flooding of the pit.
Sprinkler system installations are common in pit and crane systems and are
extremely important in facilities that store refuse on the tipping floor. The tipping bay
area should be provided with an automatic sprinkler system covering the first six meters of
the tipping floor adjacent to the refuse pit (387), and an automatic sprinkler system should
be provided over the entire refuse pit area (including the charging hoppers and the crane
park area at both ends). A minimum water density of 10 l=min=m2 should be provided
(388). This provides a discharge rate of at least 62 l=min=m of pit opening. In plants where
the crane operator rides in a cab on the bridge, care should be given to the facility design to
assure easy and rapid escape from the cab, noting that the upper regions of the building
will rapidly fill with dense, acrid smoke during a pit fire.

C.

Bin Storage and Reclaim Systems for RDF

Systems to store and retrieve RDF are critical to the operational success of all classes of
RDF combustion systems. Because of the unique physical properties of RDF, experience
has shown the importance of dynamic storage. That is, processed RDF, sitting quiescently
in a bin or other storage bunker, has a tendency to knit together into a coherent structure
that, often, will not behave in a predictable and desirable manner when outtake conveyors
or another withdrawal mechanism is started. For example, it may be found that, after a
period of time, a conveyor mounted under the RDF pile simply bores a hole into the mass
and the structure overhead fails to collapse into the void. Thus, after only a small quantity
is withdrawn, the RDF flow rate drops to zero.
Dynamic storage involves both continuous operation of the withdrawal mechanism
and continuous recycle of the unfed RDF flow. Therefore, when RDF is not being fired,
there is 100% recycle. Then, as the firing rate increases, the recycle stream diminishes but
does not stop. A second aspect of successful dynamic storage has sometimes involved the
installation of vertical screws in bins to continuously fluff the RDF. Parallel paired
counterrotating withdrawal screws in the bin bottom provide a live bottom feature that
also has proven successful. Note, however, that continuous movement exacerbates the wear
problems fostered by the highly abrasive nature of waste and its accompanying sand and
grit. Also, rotating elements (such as the screws) become fouled by ribbon-type waste
materials.
The RDF storage bin is important because, in most instances, the RDF preparation
system is only operated on a one- or two-shift basis. As noted above, the RDF processing
capacity is often considerably greater than the needs of the combustion facility. Thus, a
facility is needed to hold a relatively large working volume of waste to bridge over the shift
outages and maintenance outages of the processing system.
Ideally, a surge storage bin is interposed between the long-term storage bin and the
combustor. The quantity of material in the surge bin usually corresponds to 5 to 10 minutes
of firing at the design rate. This bin should respond quickly to changes in boiler demand.

V.

SIZE CONTROL AND SALVAGE

In some plants, oversize or otherwise undesirable refuse is controlled by exclusion at the

tipping location or by selective removal from the pit by the crane operator. Under
supervision, bulky articles can be deposited in a specific area of the pit, or they can be
carried directly to the residue disposal site without passing through the incinerator furnace.
The latter method is often used for bulky metal items such as refrigerators, stoves, and
large metal parts (white goods) that contain little or no combustible material. If the item
contains biodegradable material, it may be cleaned before it is taken to the residue disposal
site. Combustible oversize items are preferably broken up and incinerated.
Most refuse can be reduced in size by shearing or shredding. Shearing is a controlled
method for reducing all refuse to a given maximum size by passing it through a shear or
series of shears, or through a single or multiple guillotine. Shredding, which is rapidly
coming into increased use, commonly makes use of horizontal or vertical shaft hammermills or shear shredders. Coarse shredding (say to a top size of 25 cm) improves the
efficiency of recovery of ferrous metal with magnetic drums or belts. Finer shredding,
although necessary for applications where the refuse will be burned in suspension, presents
problems when burning on conventional incinerator grates.

VI.

INCINERATOR FEED SYSTEMS

A.

Feed Systems for Floor Dump Receipt and Storage

Most floor dump receipt and storage systems use a front-end loader to reclaim refuse from
the storage piles and deposit it into feeding hoppers. The front-end loaders are operated
continuously to feed waste into the incinerators. The front-end loader bucket should have
sufficient capacity to fill the waste feed hopper completely with one load. The bucket
capacity should have a minimum volume of 1 m3 to accommodate the maximum
dimension of typical wastes but must be sized to be compatible with the incinerator unit
size. Because of the high dust level on the charging floor, the radiator on the front-end
loader should be extra heavy-duty. Also, the tires should be solid rubber. Since the loader
is critical to continued plant operation, the plant should either have two loaders on-site or
have arrangements with local heavy equipment firms to assure reliable and fast delivery of
rental vehicles.
The refuse is ultimately recovered from the piles (last in, first out) and deposited at
the same grade as the storage floor. This design approach is simple and is the easiest to
work with, but it requires that the elevation of the tipping floor must be considerably higher
than the floor of the incinerator system. This may present problems with the site layout but
is greatly preferred from an operational standpoint. The refuse is then
Pushed onto a conveyor that loads the feed hopper of an incinerator furnace or a
shredding or compacting device.
or
Pushed through an opening in the floor either to directly feed an incinerator furnace
or to load a hopper equipped with a hydraulic ram for feeding to an incinerator
furnace in a batchwise fashion. There are two closures on the hopper to ensure
that there is always a seal between the incinerator environment and the tipping
area. The seal limits or stops the flow of cold air into the furnace (usually under
draft) on charging. An inrush of cold air is detrimental to the inside refractory

walls of the furnace and can cause CO and smoke by cooling and quenching the
burning process.
The first closure is a hydraulically actuated, reasonably airtight top closure to the
charging hopper cavity (a lid) that closes to isolate the hopper from the tipping
floor after the waste is fed into the hopper. The second closure is a guillotine
damper that opens to allow the hydraulic ram to push the waste from the hopper
into the incinerator. The face of the guillotine door that is exposed to the furnace
is refractory-protected. After the charge of waste is pushed into the furnace, the
feed ram retracts and the guillotine closes to isolate the hopper from the
incinerator. The hopper top closure then is opened to allow the hopper to be
reloaded with waste from the tipping floor for another feed cycle.
In a few cases, the waste must be scooped into the bucket of a front-end loader and lifted to
the dumping point. Such handling should be avoided if possible since it is slow and,
including the cleanup of spillage, labor-intensive.
The rate of feed is controlled by a time cycle or plant instrumentation that indicates
more waste is required. For example, modular, starved-air incinerators operate with a
fixed air flow to the primary chamber. The flow is low and the chamber is maintained in the
reducing mode (with less than a stoichiometric oxygen supply). As the combustible burns
down, the air supply comes into better balance with the heat release and the flue gas
temperature increases, thus signaling the need to charge new waste. In many plants a light
located directly above the incinerator-loading hopper indicates to the front-end loader
operator that the hopper is empty and positioned to receive a load of waste. The waste feed
cycle time varies depending on the design capacity of the incinerator. Typically, waste is
fed into the hopper every 2 to 4 minutes.
With a hopper-feed system, either the ram can clear the hopper at each stroke or an
oversized hopper can be filled with refuse and the ram used to shear a horizontal section of
refuse at selected intervals. The ram feeder provides an air seal at the feed to the furnace,
an improvement over the bucket or the front-end loader systems of batch feeding, which
often let in undesirable quantities of cold air, as well as releasing occasional puffs of flames
or hot gases when the charging gate was open.
B.

Feed Systems for Pit and Crane Receipt and Storage Systems

1.

Mass Burn Systems

Modern-design mass burn incinerator systems nearly always specify continuous or

semicontinuous feeding of refuse to the incineration furnace (Fig. 3). In larger mass
burning plants, a hopper and a gravity chute supply a mechanical feeder, such as a pusher
ram. RDF systems use a mechanical or pneumatic conveyor to transport the refuse from a
storage bunker and then continuously cast the RDF into the furnace using an air-swept
chute and high-pressure jets of air.
In the hopper and gravity chute systems (Fig. 4), a rectangular hopper receives
refuse delivered by the crane and bucket. The bottom of the hopper terminates in a
rectangular section chute leading downward to the furnace grate or other feeder conveyor
at the entrance to the furnace chamber itself. This gravity chute is either of uniform crosssection or flared slightly toward the lower end to minimize bridging. It is either refractorylined, made of water-cooled steel plate, or both, to prevent heat damage when the
occasional burn-back occurs or at shutdown. The inside walls should be carefully

Figure 3 Continuous-feed incinerator.

Figure 4 Detail of hopper and chute system. (Courtesy of Von Roll, Inc.)

ground and otherwise smoothed to avoid even small irregularities and ledges that will
catch refuse components and hold back the flow of refuse. Fires in the chute may be caused
by faulty chute design (too short or too narrow), by the presence of large or irregular
refuse, which can provide flues for combustion gases and flame to ascend the chute, or by
having insufficient material in the chute.

2.

RDF Burning Systems

Floor dump and pit and crane=grapple feed systems have the relatively simple and
predictable task of moving a unit volume of raw waste from a pile or pit to the charging
hopper. However, RDF handling technology must go far beyond this and must effectively
and reliably master the vagaries of waste physical form (ribbons, massive chunks,
powders, wet sludge, etc.) and provide the combustion system with a relatively uniform
energy feed rate (kcal=min).
Selecting equipment for storage and retrieval of RDF is an inexact process. The
designer must be cognizant of the physical and chemical characteristics of both the RDF
and the original raw feed. Of special importance are the handling problems (hangups,

choking, etc.) from rags, streamers, wires, hard plastics, and metals. Also, one must
recognize that RDF is not uniform. Moisture content will vary, particle size will change,
unexpected materials will appear from time to time. RDF can be fluffy or dense and is
relatively incompressible. Although the RDF preparation system is supposed to remove
many of the problem materials and minimize RDF variations, the expectation should be
that this protection will fail from time to time. When it does, the consequences will exact a
toll on the steadiness of feed (and, thus, the combustion and steam generation rate) and the
distribution to the several feed spouts (and, thus, the distribution of RDF on the grate), etc.
Unlike the mass burn systems, heat release (and steaming rate) within RDF
combustors tracks much more closely the energy feed rate to the system. Although
there is a certain amount of heat release inertia from the mass of material burning on the
grate, an estimated 20% to 30% of the heat is released in the suspension burning part of the
two-step combustion process. This close coupling of feed rate with steaming rate
emphasizes the importance of designing hardware that minimizes stopstart flow, frequent
blockage of one or more feed points, etc. Experience to date has often shown that the
design process involves substantial trial and error. Success is seldom achieved in the early
phase of plant startup and shakedown.
a. Spreader Stoker (RDF) Burning Systems. A spreader stoker system consists of a
feeding device, a mechanism for throwing the RDF into the furnace, and a simple flat
traveling grate with suitable openings through which to admit underfire air. In some
designs tailored to handle high-moisture, high-ash wastes, a reciprocating, water-cooled
grate is used. The RDF feeding and distributing mechanism is located at the front wall of
the furnace (usually a waterwall boiler enclosure), above the grates. The grates move
toward the feeder assembly and discharge the ash just below the front wall. The speed of
the grates is set to give sufficient time for combustion to be completed before the ash is
discharged at the front. The residue is continuously discharged from the furnace into a
hopper. Normally, the ash is relatively cool and clinkerfree.
After injection, the RDF transits the volume above the burning material on the grate.
The hot, still-burning gases rising from the grate pass through and about the stream of
entering waste. As the flue gases sweep past the incoming waste fragments, the waste dries,
ignites, and begins to burn. The burning waste stream then falls to the grate for burnout.
This two-step combustion process is sometimes called semisuspension burning.
The feed system usually begins with a conveyor system drawing from the surge bin
that may incorporate a weigh feeder device to meter the mass rate of RDF. In most cases,
the materials handling system is a run-around design where the mass rate of material
withdrawn from storage is significantly larger than the net feed rate. The excess is returned
to storage. This strategy helps to minimize irregularities in the feed rate. The conveyor
discharges into a distribution device that divides the feed into the 2 to 8 chutes leading to
the feeding devices. Alternatively, the system can be designed with multiple discharge
points from surge bins interposed between the large, long-term storage RDF holding bins
and the boiler. Whatever the design, the feeder design must provide for easy and rapid
access for routine inspection, for preventive maintenance, and for operator intervention to
break up bridging and other flow-interrupting problems.
The spreader stoker feeder functions to vary the supply of RDF to the furnace and to
provide even distribution on the grate. The mechanisms that propel the RDF into the
furnace include steam and air injection as well as mechanical rotors. Mechanical spreader
stoker feeders are divided into two classes: overthrow, where the RDF comes into contact

with the upper part of the rotor blade assembly, and underthrow, when it comes into
contact with the lower part.
Pneumatic feeders are most often used with RDF. A widely used technique for
distribution of RDF uses a swinging spout that services multiple feed chutes. The RDF is
fed into the top of the unit and the swinging spout proportions the RDF into 2, 3, or 4
chutes running to the feeders. Rollers at the nexus between the chutes avoid hangup of
ribbons and rags and break up bridges.
Figure 5 shows a pneumatic system used for RDF firing. Air, under pressure, enters
the distributor inlet under control with a damper. The direction of the air flow in the
horizontal plane is set with adjustable air vanes in the air distributor nozzle. Waste, falling
by gravity from a distribution system above, is caught up in the powerful air jet and swept
toward the furnace opening. The final fuel trajectory is controlled by angling the distributor
tray by means of the distributor tray control. Figure 6 shows the boiler system, including
the pneumatic feeder and traveling grate stoker.
b. Suspension (RDF) Burning Systems. If suspension burning is to be employed in the
incinerator furnace, the refuse should be prepared by suitable shredding or grinding. The
most desirable feeding method is air injection. Suspension burning has been used
successfully in waterwall boilers for the generation of steam from waste products such
as wood bark, bagasse, and similar materials. In conformance with practices in the fossil
fuel-fired boiler industry, air injection of shredded refuse has been used for corner firing,
for spreader stoker firing, or for cyclone firing, all of which are in commercial use for
the generation of steam from powdered or crushed bituminous coal. With suspension
burning, a burnout grate should be provided at the bottom of the waterwall furnace
chamber for completion of combustion of large particles and=or slow-burning materials.

Figure 5 Pneumatic refuse fuel feed system. (Courtesy of Babcock Borsig Power, Inc.)

Figure 6 Pneumatic RDF fed boiler system. (Courtesy of Babcock Borsig Power, Inc.)

VII.

GRATES AND HEARTHS

In general, small incineration furnaces use a stationary grate or refractory hearth to support
the burning refuse. Most larger plants use one of several available grate types to support
and transport the refuse while simultaneously stoking or mixing the refuse during the
combustion process. Suspension burning is the only process that does not necessarily
require a hearth or grate, since most of the refuse is oxidized while in suspension in
furnace gases. But even for suspension burning, a burnout grate is often provided at the
bottom of the furnace chamber to provide for more complete burnout of combustibles in
the residue. Experience suggests that only when the refuse has been shredded to 95%
< 1 cm should consideration be given to abandoning the burnout grate.
There are many different types of hearths or grates, each of which has its own special
features.
A.

Stationary Hearth

Those incinerator furnace systems that operate without grates include the stationary hearth,
rotary hearth, and rotary kiln. The stationary hearth is usually a refractory floor to the
furnace, which may have openings for the admission of air under slight pressure below the
burning material on the hearth, in the manner of the blacksmiths forge. In the absence of
underfire air ports, air is admitted along the sides or from the top of the furnace, and
combustion proceeds in the same, surface-burning manner as in a bonfire, but with
improved conditions due to the reradiation of heat from the surrounding furnace walls and
roof. Unless the refuse being processed provides a porous burning mass (as may be the
case with bulky refuse), it is necessary to provide manual stoking with slice bars to stir the
mass of refuse in order to achieve a reasonable burning rate and degree of burnout.
Stationary hearth furnaces are used for most commercial and smaller industrial
incinerators. They are also used almost exclusively in crematories and for hospital wastes.
For the latter applications, auxiliary gas or oil burners are used to maintain the furnace
temperature in the range of 650! to 900! C in the presence of adequate quantities of oxygen
from air, well dispersed throughout the gas. In burning such high-moisture waste, primary
chamber burners are needed to ensure complete oxidation of combustible solids and vapors
for the elimination of smoke and odors. Confident smoke and odor control requires auxiliary
fuel burners under automatic temperature control in a secondary combustion chamber.
Bulky refuse incinerators constructed in the mid-1970s (117,118,121) used a
refractory chamber with a refractory floor equipped with a number of air inlet gratings,
similar to floor drains. One side or end of the chamber was connected to the hot
combustion chamber of another incinerator installation. If not, a secondary combustion
chamber was installed prior to the breeching and stack to ensure burnout of combustible
gases and particulate. A refractory door permitted the opposite side of the combustion
chamber to be opened for access. A special extended bulldozer blade was used to load the
hearth with refuse from a paved platform outside the chamber and to drag out ash residue.
Alternatively, it was possible to remove the ash residue by raking with a hoe or by pushing
the ash residue beyond the hearth into a depressed area from which the ash could be
removed.
B.

Rotary Kiln

The rotary kiln hearth system has been used for several hundred years in the pyroprocessing industry. The rotary kiln concept accomplishes two objectives simultaneously:

moving solids into and out of a high-temperature combustion zone and mixing and stoking
the solids during combustion. In its classical embodiment, a kiln is comprised of a steel
cylindrical shell lined with abrasion-resistant refractory to prevent overheating of the
metal. A kiln is generally inclined slightly toward the discharge end, and the movement of
the solids being processed is controlled by the speed of rotation.
In some instances, special provisions have been made for air cooling or water
cooling the kiln shell, thus eliminating the use of refractory. For example, a steel kiln
cooled with external water sprays known as the Lantz Converter was developed for
pyrolysis applications in the mid-1960s. In the late 1970s, a new kiln technology emerged
involving a water-cooled kiln cylinder fabricated of steel boiler tubes welded together and
manifolded for the introduction of feedwater and withdrawal of steam. This system (The
OConnor Rotary Combustor) is described in Section F. The design and operating
features of kilns for the destruction of hazardous wastes are described in Chapter 11.
There has been little use of the refractory-lined rotary kiln for municipal incinerator
furnaces except to enhance residue burnout after the burning of refuse on a multiple-grate
system. The rotary kiln normally requires all of the air for combustion to enter the unit
with the refuse at the feed end. Therefore, the cooling flow of ambient combustion air is a
deterrent to the temperature increase necessary for rapid ignition. Further, the inherent
construction of a kiln emphasizes the suspension of particulate in the gas stream and tends
to limit mixing of pyrolyzed refuse combustible matter with combustion air. For kiln-based
incinerators, these features result in high particulate and hydrocarbon concentrations in the
flue gases entering the APC system in comparison with conventional grate systems. The
few rotary kilns that have been used for incineration of refuse have been relatively stubby
(length limited to only about four to five times the diameter) and are commonly provided
with a bypass duct to reduce the gas volumetric flow rate (and particulate entrainment)
through the kiln.
C.

Stationary Grates

Stationary grates have been used in incinerator furnaces for a longer time than any grate
system except the stationary hearth, which was used in incinerators and crematories prior
to the Middle Ages. The original stationary grates were probably metal bars or rails
supported in the masonry side-walls of the furnace chamber. Subsequently, these bars were
replaced with cast metal or fabricated metal grates with mechanical linkages to rotate the
grate sections to dump the ash residue. While some stoking action can be obtained by
shaking the grate, as in partial dumping, stationary grates normally require manual stoking
with a slice bar to stir the burning bed of refuse in order to obtain a reasonably complete
burnout of ash residue. Such stationary grates are not in wide use in modern facilities but
were common in many of the older cell-type incinerators. The latter have usually been
constructed with a multiplicity of furnace sections or cells, with an opening above each cell
for charging.
D.

Mechanical Grates: Batch Operations

Mechanically operated grates installed in batch-type furnaces were a natural evolution

from the stationary grate furnaces. Although batch-type incineration furnaces have given
way to continuous furnaces for new, large installations, many of the new small-capacity
incinerators still utilize batch-fed furnaces, either with stationary or intermittently operated
grates or without grates, the latter in small commercial and industrial installations.

1.

Cylindrical Furnace Grates

Cylindrical batch furnaces were common in the United States in the 1950s and early
1960s. In these furnaces, the grates form annuluses inside the vertical cylindrical walls of
the furnace. A solid grate or dead plate covers the central area of the annulus. A hollow
rotating hub with extended rabble arms rotates slowly above the circular dead plate to
provide mechanical stoking or mixing. A center post provides the bearing support of the
hub and rabble arms. The rotating hub is covered with a hemispherical cone with one or
more consecutively smaller cones stacked on top of the first one. Forced air (called cone
air) for combustion is supplied through the hub to the hollow rabble arms and then
through openings in the arms to the space just above the dead plate. Additional cone air is
supplied to each of the cones, to cool the metal. The annular grate area is divided into pairs
of keystone-shaped segments; each pair is arranged to open downward for dumping the
ash residue into the ash hopper below. These segmental grates are either hand-operated or
hydraulically operated under manual control. It is frequently necessary to poke and slice
the ash residue using the access doors around the furnace, in order to clean the fires and to
assist in the dumping of the ash residue.
When circular grates are used, the cylindrical furnace is fed intermittently or
batchwise through an opening in the top at the centerline of the furnace. A charging
door is used to close the opening between additions of refuse. In the usual method of
operation, refuse is placed in the hopper above the charging door or gate preparatory to the
next charging cycle. At a signal from the operator, the charging gate is opened, the refuse
drops into the bonfire of previously ignited refuse, and the charging gate closes
immediately after the hopper is emptied. Additional refuse is charged in repeated cycles
at the discretion of the operator until the accumulated ash residue in the furnace requires
dumping through the circular grates.
2.

Rectangular Batch Furnace Grates

Mechanically operated grates in rectangular batch-operated incinerator furnaces include

reciprocating (pusher) grates, and rocking grates. The grates are usually installed in a
slightly inclined position from the horizontal, with the lower end of the incline at the ash
discharge point. With these grates, the furnace is fed intermittently through an opening in
the top and at the higher end of the grate, and the fresh refuse is deposited over the bonfire
of previously ignited refuse. These systems were contemporaries of the cylindrical
furnaces in the United States during the 1950s and 1960s.
As the burning continues, the grates are operated under manual control to move the
burning bed of refuse toward the discharge; the manual control is to prevent (ideally) the
discharge of residue that has not been completely burned. In some instances a dump grate
is installed at the ash discharge point to hold back ash residue that is still burning, with
manual operation of the dump grate after the accumulated ash has been completely burned.
At times it is necessary to manually stoke or slice the fire in order to spread the burning
refuse over the grate or remove larger pieces of metal or clinkers that interfere with the
desired flow of burning refuse across the grate.
E.

Mechanical Grates: Continuous Operations

Mechanical constant-flow grates have been and are being used in most of the newer
continuous-burning municipal-scale incinerators. This constant-flow grate feeds the refuse
continuously from the refuse feed chute to the incinerator furnace, provides movement of

the refuse bed and ash residue toward the discharge end of the grate, and does some
stoking and mixing of the burning material on the grates. Underfire air passes upward
through the grate to provide oxygen for the combustion processes, while at the same time
cooling the metal portions of the grate to protect them from oxidation and heat damage.
Typical grate designs correspond to an average heat release rate of 13,500 to 16,000 kcal
m"2 min"1 . Clearly, the actual rate in different portions of the grate differs widely from this
average.
The grate device is only one part of the overall primary furnace system. The primary
furnace system includes the hopper=chute=feeder, the grate and underfire air system, the
furnace enclosure with its protective side walls, the overfire air system (including design
features such as the shape of the lower regions of the enclosure and the direction and
discharge velocity of the air jet nozzles), and the bottom ash and siftings equipment. While
each vendor makes claims as to the benefits of its particular grate design, most firms have
invested considerable time and money in developing an optimized, overall primary furnace
system that works together as a successful, integrated package. If one takes the grate of one
vendor and inserts it into the furnace envelope of another, success is no longer as certain.
1.

Reciprocating Grate

Reciprocating grates involve cast alloy grate bars, actuated in sequence to push, mix, and
help to break up (stoke) the refuse. Since the early 1970s, the reciprocating grate system
has become the most commonly used type in mass burn incinerator service. The plane of
the grates provided by the several grate vendor firms throughout the world ranges from flat
to inclined downward toward the ash discharge point. The grate bar actuation devices
include those that push toward the ash discharge (the preponderance of grates are of this
design) and those that push toward the feed end.
a. Grates on the Flat or Pushing Downhill. In this embodiment, the reciprocating or
pusher grate is installed stepwise in rows on the flat or on a slight downward incline toward
the discharge (generally up to about 15! depression). Alternate rows of high chromium
nickel alloy grate bars are stationary or move to convey the refuse from the feed point
(under the feed chute or at the discharge plane of a feeder ram) through the combustion
area to the ash hopper or other ash discharge equipment. Stoking action comes as a result
of the separate motion of the alternate rows of grates. Additional stoking and mixing
(breaking open packed refuse masses) is obtained in some designs by providing one or
more dropoffs, causing tumbling from one grate section to the next. Two, three, or four
grate sections are commonly included in this type of grate for a continuous-flow
incinerator. This design is illustrated in Fig. 7.

b. Grates Pushing Uphill. The reverse acting reciprocating grate (Fig. 8) is also
inclined downward, though at a steeper angle, toward the discharge end. However, in this
system, the cast metal grate elements reciprocate uphill against the downward flow of
refuse, thereby producing some rolling of the burning material as a result of the upward
reverse thrust. Again, these grates form a steplike configuration rather than a steady
incline, so that additional mixing is obtained as the refuse tumbles from one level to the
next.

Figure 7 Detroit Reciprograte1 refuse fired stoker. 1, refuse charging hopper; 2, refuse charging
throat (refractory-lined or water-cooled); 3, hydraulic charging ram; 4, grates (high CrNi alloy); 5,
tapered roller bearings (shielded from siftings); 6, hydraulic power cylinders and control valves; 7,
vertical dropoff (lined with alloy tuye`res to admit air); 8, overfine air jets; 9, undergrate combustion
air supply ducts; 10, automatic siftings removal system (discharges to bottom ash system). (Courtesy
of Detroit Stocker Company.)

2.

Rocking Grate

The rocking grate slopes downward from the feed toward the discharge end. Two, three,
four, or more grate sections are installed in series. There is little or no (e.g., < 3 cm)
dropoff between sections. The rocking grate sections are comprised of a rectangular array
of grate segments each of which is approximately quarter-cylindrical. Each segment has
20% to 30% open area to pass siftings to a hopper or conveyor system mounted below the
grate and to admit undergrate air. Alternate rows of grate segments are rotated approximately 90! about the edge toward the discharge of the grates.
As the grate face rises up into the burning bed, the refuse mass is broken up and
thrust forward toward the discharge. These grate segments each rotate back to a rest
position and alternate segments rotate as before, causing a similar stoking action and
pushing of the refuse bed forward. This two-cycle action is similar to the previously
described pusher and reverse-acting grates, except that it seems to mix the material more
effectively. Further, it redistributes the refuse and ash on the grate to cover local areas

Figure 8 MARTIN1 reverse-acting grate. 1, feed hopper; 2, feeder; 3, MARTIN1 reverse-acting

grate; 4, MARTIN1 discharger; 5, furnace; 6, steam heated air preheater; 7, injection of combustion
air (underfire air); and 8, injection of combustion air (overfire air). (Courtesy of Ogden Energy
Group, Inc.)

where burning has left some grate surface exposed. The rocking grate was common in U.S.
incinerator plants of the 1960s but rapidly declined thereafter.
3.

Vibrating, Oscillating, and Impact Grates

The vibrating or oscillating grate is mechanically powered by an eccentric-weight vibration

generator or an eccentric-driven connecting rod. The oscillations convey the refuse
through the incinerator furnace from one grate section to the next, in the same manner
as an oscillating or vibrating conveyor. A similar, but nonuniform, oscillation cycle is used

for the impact stoker or grate. The grate sections are moved either forward or in reverse
and then released for spring return against a stop (impact). The grate surfaces are generally
of the same type as used in the reciprocating grate elements and indeed, might be
considered as a variation of the reciprocating type of grate. Although oscillating and
impact grates have been used in fuel-burning furnaces and incinerators for many years,
there are fewer of them than the other types of grates thus far described.
Figure 9 shows the oscillating grate system as used to fire biomass fuels (including
RDF) that vary in moisture content and heating value. The systems shown uses watercooled grates for reliability and protection, even when firing the boiler with auxiliary
burners. Boiler feedwater is used to cool the grate, thus recycling the recovered heat to
steam generation. The grate is fabricated of high-temperature cast materials, seated on
water-cooling tubes with high-conductivity grout. Siftings fall into the plenum and are
carried to the discharge opening by the grates oscillating motion. Only a small (2- to 4horsepower), variable-speed drive motor is needed to drive the stoker oscillations.
Typically, oscillation cycles are approximately 5 min apart with oscillation periods of 5
to 10 sec and an oscillation amplitude of only about 0.6 cm.

Figure 9 Oscillating grate system. (Courtesy of Babcock Borsig Power, Inc.)

4.

Traveling Grate

The traveling grate was the most widely used grate for continuous-flow mass burning
incinerator furnaces through the early 1970s. It remains the most common alternative in
RDF-fired furnaces. There are many installations in the United States, in Europe, and in
Japan. The traveling grate has been in use for many years in coal-fired furnaces and, as in
most grate systems, was adapted for use in municipal incinerators. There are two types of
traveling grate stokers: the chain grate and the bar grate. Both convey the refuse fuel from
the gravity feed chute through the incinerator furnace to the ash residue discharge, much as
a conveyor belt.
In RDF applications, grate heat release is approximately two million kcal=m2 -hr,
with about 12.8 kcal=hr heat input per meter of grate width over 45% to 50% of the grate
width. Typical grate speeds approximate 7.6 m=h.
Figure 10(a) shows a partially assembled, spreader feeder-fired traveling grate
showing the drive sprocket, skid shoes, and grate racks. As the clips that comprise the
exposed grate surface travel around the sprocket ends, they separate to free trapped ash.
Figure 10(b) shows the assembled stoker as it rounds the drive shaft. Note that the grease
lines penetrate the side wall so that the grate mechanism can be completely lubricated
while the unit is in service.
Because the traveling grate stoker does not stoke or mix the fuel bed as it conveys,
mass burning incinerator traveling grate stokers are often cascaded in two, three, and even

Figure 10 (a) Internal features of a traveling grate system near the drive sprocket. (b) Drive
sprocket of a traveling grate system. (Courtesy of Babcock Borsig Power, Inc.)

Figure 10 Continued.

four or more units. With two units in cascade formation, the first portion under the feed
chute and continuing through the ignition zone is often called the feeding or drying stoker.
The discharge from this first portion of the traveling grate spills and tumbles onto a second
unit, thereby mixing, breaking, and redistributing the fuel bed on the second stage. With
three or more stages, this spillover and redistribution are repeated. Since the thickness of
the refuse fuel bed decreases as the combustion process continues, the linear conveying
speed of successive traveling grate units is adjusted to accommodate the rate of
combustion and to provide more complete burnout after the combustion rate has subsided
(decreased linear speed).
Attempts have been made to install auxiliary mixing devices over the fuel bed on the
traveling grate. However, these have not been successful, mainly because of entanglements
with scrap wire and other debris contained in the refuse. The best method for stoking or
mixing with the traveling grate is to use several grates so that stoking and redistribution are
accomplished as the burning refuse is spilled from one section to the succeeding section.
5.

Drum Grate

The drum grate, or roller grate (sometimes called the barrel-type grate), was developed by
the city of Dusseldorf, West Germany, and subsequently commercialized. The intent of the
concept was to reduce the cost of installing multiple traveling grates to obtain better

burnout. Each rotating drum or cylinder represents a minimum length of a traveling grate
section, thereby providing a maximum number of tumbling zones in the fuel bed. This
rotating grate system consists of six to eight or more rotating grate cylinders, 1.5 m in
diameter, mounted on a horizontal axis in close proximity and at a downward inclination of
about 30! .
The slow rotation of the drums creates a strong mixing action in the refuse between
successive drums, thus loosening and opening bundles of paper or plastic bags. The
rotating cylinders extend over the width of the furnace (width of the grate), and each
cylinder is equipped with its own variable-speed drive to control the movement of the
burning refuse through the incinerator. Thus, the rotating drum grate provides better
control of refuse stoking and residence time during combustion than the traveling grate
with a consequent improvement in burnout.
F.

OConner Rotary Combustor (Kiln)

The OConner Rotary Combustor is a relatively new combustion concept. The rotating,
cylindrical kiln is fabricated of boiler tubes much as a waterwall panel, but it incorporates
perforations in the web between tubes. The kiln tubes, which absorb from 25% to 35% of
the heat released from the refuse, terminate in a torus-shaped header at the feed and
discharge ends of the kiln. A spider of feedwater inlet and outlet piping springs from a
rotary joint at the center of the discharge. The kiln discharges into a rectangular waterwall
radiant boiler. The waste enters at the high end of the kiln through a choked chute and
hydraulic ram and proceeds down the kiln. Below the kiln discharge is a steep,
reciprocating burnout grate terminating in a water quench tank and hydraulic ram ash
discharger or similar device.
The kiln rides on tires and trunnion bearings as a conventional kiln. However, the
kiln is surrounded by a segmented windbox. The pressurized segments of the windbox are
isolated using flexible metal wiping seals that rub against a plate on the kiln. Air is
supplied to the combustion volume from plenums both under the waste (underfire air) and
above the fire (overfire air) through the perforations between the boiler tubes comprising
the kiln structure.
The rotary combustor design speaks to several of the problems inherent in refractory
kilns. It is water-cooled so that rapid heat release does not lead to hot spots and clinker
formation. The air supply system allows air to be introduced along the flow path, not just at
the feed end. However, the air injection means is relatively ineffective in mixing the gases
and in destroying the natural stratification common to kiln devices. Thus, the carbon
monoxide concentration in the gases leaving the primary furnace zone and arising from the
burnout grates is somewhat greater than is found in many other systems. Secondary air
injection in the waterwall secondary chamber is designed to address this situation.
G.

Fluid Bed Systems

The general characteristics of fluid bed (FB) incineration systems are presented in Chapter
9 relative to bubbling fluid bed (BFB) applications in burning biological wastewater sludge
and in Chapter 11 relative to circulating fluid bed (CFB) applications in burning hazardous
wastes and will not be repeated here. These two waste management areas were the
dominant applications of fluid bed technology in the United States at the beginning of the
1990s, but several applications for municipal solid waste could be found in Japan and
Europe.

In order to provide a feed with relatively homogeneous composition and size, almost
all solid wastes fired in BFB or CFB units are preprocessed or, at least, made somewhat
more homogeneous through waste separation at the point of generation (e.g., in Japan).
Both BFB and CFB furnaces can accept other fuels such as wood chips, coal, or chipped
tires. By acquiring these relatively clean, alternate energy sources when the prices are
favorable, energy revenues can sometimes be increased significantly with little increase in
maintenance and operating cost. BFB applications are discussed in Section XIII of this
chapter.
Over 32 plants in Japan (383) make use of a fluidized bed process that, in a sense, is
halfway between grate systems and CFB: the revolving fluidized bed (RFB). The
characteristics of the process favor smaller units (say, 60 to 130 tonne per day). In the RFB,
the waste is burned without fine preshredding. Only rough tearing in a twin-screw feed
mechanism is applied. Note, however, that items of furniture, whole bicycles, or
mattresses, etc., cannot be fed without processing. The units have been operated with
high-plastic (high heat content) waste, ordinary refuse, and 79%-moisture-content sewage
sludge or night soil (with auxiliary firing). Co-incineration of sludge and refuse is
practiced in several facilities (383).
The RFB is fed through the freeboard. The feed falls into a vigorously moving bed
with a constriction (the deflector), which reduces the cross-sectional area of the furnace
to about 40% of the distribution plate plan area. This leads to both vertical and lateral
rotation of bed sand and of burning refuse in the volume between the distribution plate and
the plane of the constriction. The lateral motion is aided by a peaked conical shape to the
distribution plate that directs noncombustible to water-cooled screw conveyor discharge
points on the periphery. Bed temperatures are maintained between 700! and 800! C.
Data on NOx from RFB systems show less than one-half of the concentrations found
in grate systems (383). This probably reflects both lower mean and peak temperatures and
some in-bed reduction.
In comparison to grate-burning process, FB systems offer intense mixing, long
residence time (typically, 4 to 5 sec), excellent residue burnout (typically, less than 1%
unburned carbon), lower combustion temperatures (less thermal NOx , higher boiler
efficiency, and less flue gas volume for air pollution control due to reductions in excess
air to 40% to 90%), and lowered potential for ash slagging. Note, however, that if slagging
occurs, the adverse impact on a fluid bed system is profound. Limestone can be added to
the bed of FB units to achieve a nominal reduction in sulfur dioxide (40% to 50%
reduction) and some HCl reduction.
VIII.

INCINERATOR FURNACE ENCLOSURES

The furnace enclosure provides a controlled environment for the combustion process in the
incinerator system. Without the furnace enclosure, the combustion process would be, in
effect, open burning. If one simply installs a metal shield to protect the open burning fire
from wind and direct radiation to the atmosphere, we have the conical or tepee burner,
which was used up until the early 1970s for municipal refuse and wood wastes (Chapter
12). However, the conical unit has the technical limitation that it must be operated with
300% to 600% excess combustion air to prevent catastrophic heat damage to the metal
structure and enclosure. At this level of dilution, the quantity of flue gases generated and
their flow rate variations are so extreme that air pollution control is untenable. In its time,
however, the tepee provided a working example of the most basic concept in furnace
enclosure.

Materials for conventional incinerator furnace enclosures (see also, Chapter 5)

include
Nonmetals: refractories, e.g., firebrick walls and roofs
Metal: plate, tubes, pipe, etc., fabricated of single metals or clad metals offering a
cost-effective combination of cost and corrosion resistance
Refractory-covered metal: castable or firebrick refractory lining or coatings, 3 to
25 cm thick
Either water or air may be used to cool these enclosure materials. Cooling water can be
either contained in tubes or pipes or uncontained in the form of a film on the surface of the
external metal plate. Air cooling can be employed with forced convection, with forced air
jets impinging on the surface, or by radiation and natural convection to the ambient
environment. Warmed cooling air can be used as preheated air for combustion or can be
ducted for building heat.

A.

Refractory Enclosures

1.

General

There are at least three types of refractory enclosures: gravity walls with sprung arch roofs,
suspended walls and suspended roofs, and refractory linings supported directly by the
metal shell.
Gravity walls consist of refractories laid in courses, one above the other, supported
on a foundation floor or other support, and are vertical (or almost vertical). The
sprung arch used with gravity walls is a portion of a circular arch of refractory
shapes. The arch is supported by compression of the side walls through skewbacks retained with structural steel members held together with tie rods.
Suspended refractories are supported on structural steel either directly with clamps
attached to each piece of refractory or with intermediate refractory pieces keyed
or cast into the refractory roof being supported, with the refractory supporting
member held with a metal clamp to the structural steel.
Refractory linings are installed in a circular flue or duct and held like staves in a
barrel, or if cast refractory is used, it is supported with metal hooks or dowels
fastened to the shell by bolting or welding. Refractory walls can also be cast over
pipes or steam tubes, thus protecting the metal tubes against corrosion or flame
impingement, and at the same time cooling the refractory surface, with the tubes,
in turn, cooled with hot water or steam.
The rotary kiln furnace concept is lined with castable refractory or with large brick shapes
called kiln blocks. It is also possible to use an unlined rotary kiln cooled externally with
air or water, or additionally cooled with a water film on the inside of the cylinder. This
latter concept has been applied only in industrial incineration.
2.

Refractories for Small, Multiple-Chamber Incinerators

The minimum specifications for refractory materials used for lining the exterior walls of
multiple-chamber incinerators (127) are the following:

For waste of 2800 to 4500 kcal=kg higher heating value, use

Firebrick: high heat duty, pyrometric cone equivalent (PCE) not less than 32.5
Castable refractory: > 1920 kg=m3 , PCE > 16
For wood, sawdust, and other high-temperature service, use
Firebrick: superduty, PCE not less than 34
Plastic: PCE > 34; density > 2040 kg=m3
Minimum refractory thickness for lining exterior walls (including arches) of incinerators
burning all classes of refuse are the following:
Up to and including 160 kg=hr capacity:
Firebrick: 11 cm
Castable refractory or plastic: 10 cm
Above 160 kg=hr capacity: all refractories, 23 cm
Stacks should be lined with refractory material with a minimum service temperature of
1100! C. In low-capacity units the minimum lining thickness should be 6.5 cm; in units
larger than 160 kg=hr, 11 cm.
Doors should be lined with refractory material with a minimum service temperature
of 1550! C. Units smaller than 45 kg=hr should have door linings of 5-cm minimum
thickness. In the size range of 45 to 160 kg=hr, the linings should be increased to 7.5 cm. In
units with capacities from 160 to 450 kg=hr, the doors should be lined with 10 cm of
refractory. In units of 450 kg=hr or more, linings should be 15 cm.
The thickness of refractory lining and insulation in the floor of a multiple-chamber
incinerator depends primarily on its physical location. Incinerators that are installed on
their own concrete foundations outside buildings should have 7.5 cm of firebrick lining
backed by a minimum of 3.8 cm of 1100! C insulating material. Incinerators of the
semiportable type should have sufficient air space provided beneath the incinerator so that
no damage to the pad will result. When incinerators are installed within buildings, it is
extremely important that provision be made to prevent damage to floors and walls of the
building. Such damage can be prevented by providing air passages beneath the incinerator
and adjacent to the building walls to prevent excessive heat from actually reaching the
structure. If an air space beneath the incinerator is impractical, then additional insulation
should be provided.
For incinerators with capacities up to 225 kg=hr, 11 cm of firebrick and 6.5 cm of
insulation should be provided on the floor of the mixing and final combustion chambers.
For incinerators with capacities of 225 to 900 kg=hr, 11 cm of firebrick backed by 10 cm of
insulation should be provided.
Units in high-temperature service should be provided with an insulating air space of
7.5 to 10 cm between the interior refractory and the exterior steel. This will reduce the
temperature of the refractory and extend its life. Adequate openings above and below the
incinerator should be furnished for air to enter and exit freely. In some cases forced
circulation of the air in this space may be required. Some incinerator designs utilize forced
preheated air as secondary combustion air.
The minimum thickness of interior refractory dividing walls generally follows that
required for the exterior walls. The bridge wall, with its internal secondary air distribution
channels, requires greater thickness. The minimum width of refractory material between

the secondary air channel and the ignition or charging chamber should never be less than
6.5 cm for very small units; 11 cm for units up to 110 kg=hr; and 23 cm for larger units.
3.

Refractories for Large Incinerators

The choice of refractory type and the details of installation for large municipal incinerators
are still evolving. Table 1 indicates the more common choices for placement in the
different regions of the incinerator. In general, suspended construction is preferred, to
reduce the initial cost and to reduce maintenance expense.
B.

Other Enclosure-Related Design Considerations

1.

Air In-Leakage

Refractory walls and roofs can permit air to leak in or furnace gases to leak out depending
on the draft conditions in the furnace. Airtight metal casings will minimize this problem.
Infiltration is particularly common with suspended refractory construction. It can be a
benefit in providing air cooling of the wall or roof material, with in-leakage of air adding to
that required for combustion. In many installations, however, the in-leakage of air can be
excessive to the point that furnace temperatures are reduced, combustion of the furnace
gases is quenched, and air pollution control systems are overloaded.
2.

Shape Factors

Combustion engineers know that a hot, well-mixed system supplied with sufficient air will
achieve complete burnout of the most refractory organic compounds in only a fraction of a
second. For conventional fuels, burner designs can be honed with the combined guidance
of experiment and theory (the latter greatly aided by the regularity of the system) to coax
superior combustion performance over wide ranges of operating conditions. The features
of the combustion enclosure in this instance are, therefore, more driven by issues of cost,
heat transfer optimisation, and packaging convenience than by the combustion process.
In contrast, we know that grate burning inherently is a poor starting point for the
realization of complete combustion. The air supply is spatially and temporally irregular as
is the air demand. The gasification and heat release processes are in a state of continual
upset as the reciprocating grate bars expose new surface and as piles of refuse collapse and
fall. This chaos is in stark contrast to the humming regularity of oil, gas, and even
pulverized coal flames.
In the incinerator, the physical shape of the furnace enclosure and its appurtenances
plays a key role in achieving incineration objectives. The function of the furnace envelope
in refuse-fired combustors can be critical; guiding cold gases from the discharge grate area
back to the hotter regions where, after mixing, combustion is initiated; guiding hot gases to
energy-depleted zones for ignition and drying; guiding oxygen-rich gases to the air-starved
pyrolysis zone in the second third of the grate. The furnace shape also serves to funnel and
accelerate the fuel-rich gases rising from the gasification regions along the grate to target
zones for overfire air jet mixing. The walls constrain the flow to maintain gas velocities
high enough to overcome buoyancy-driven stratification and to avoid cold spots and dead
zones: wasted combustion volume. Also, the furnace shape, facilitated by skillful
placement of (hot) refractory, can provide reradiation to the bed: supplying heat energy
for the evaporation of moisture.
The achievement of a very high degree of combustible pollutant control in municipal
incineration systems over the period from 1975 to 1985 (including reliable destruction of

201425
201425
3701425
6501425
6501425
6501425
6502400
6502400
650870
650870
260540

Furnace walls, grate to

1.2 m. above
Furnace walls, upper portion
Stoking doors

Furnace ceiling
Flue to combustion chamber

Combustion chamber walls

Combustion chamber ceiling
Breeching walls
Breeching ceiling
Subsidence chamber walls
Subsidence chamber ceiling
Stack

Slight
Severe, very
important
Slight
Slight

201425

Charging gate

Moderate
Severe, very
important
Moderate
Moderate
Slight
Slight
Slight
Slight
None

Severe, very
important
Severe
Severe

Slight

Slagging

From Ref. 129.

Air cooling is often used with subsequent discharge of the warmed air as overfire air.

Slight
Slight
Slight
Slight
Slight
Slight
Slight

Severe, very
important
Severe

201425

Incinerator part
Abrasion

Temperature
range
! C

Table 1 Suggested Refractory Selection for Incineratorsa

None
None
None
None
None
None
None

Slight
None

Moderate
Severe

Severe

Severe

Mechanical
shock

Moderate
Moderate
Moderate
Moderate
Slight
Slight
Slight

Severe
Moderate

Severe
Severe

Severe

Severe

Spalling

Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Slight

Moderate
Moderate

None
None

None

None

Fly ash
adherence

Superduty or 1st quality

Superduty or 1st quality
Superduty or 1st quality
Superduty or 1st quality
Medium duty or 1st quality
Medium duty or 1st quality
Medium duty or 1st quality

Superduty
Silicon carbide or superduty

Superduty
Superduty

Silicon carbide or superduty

Superduty

Recommended
refractory

the precursors of dioxin and furan compounds) was a remarkable technical achievement of
the industry. Intensive studies of furnace shape, overfire air injection design, and
combustion controls (coupling combustion environment sensors to refuse feed rates and
air supply) combined to accept and meet the challenge of dealing with grate-fired systems
burning unprocessed, nonhomogeneous waste. Water table (flow table) techniques were
used to study the effect of furnace geometry on flow pattern. This design, development,
and analysis technology proved to be an important tool in the analysis of the refuse
combustion process and is used by most of the major furnace vendors. With this technique,
different furnace cross-sections and overfire air injection concepts could be studied at a
fraction of the cost of real systems. Changes could be made quickly and problems could be
detected promptly.
From the flow table studies and other work, different incinerator system firms have
reached somewhat different conclusions about the optimum furnace configuration. Each
firm began with a unique, proprietary grate design. This produced a starting point that is
particular to the firm. Not surprisingly, from these different starting points, different
endpoints emerged.
The three basic concepts in primary furnace combustion chamber geometry are
shown in Fig. 11:
The parallel flow concept, where gases from the solids feed point (the feed arch) are
guided down along the grate over the refuse, rising over the discharge area into
the radiant boiler;
The countercurrent or contra flow concept, where the gases from the discharge area
of the grate flow back over the refuse bed, rising over the ignition area into the
radiant boiler

Figure 11 Primary furnace configurations.

The center flow concept, where gases from the ignition area flow down the grate in
the direction of refuse movement until they meet the gases from the discharge
area that are flowing back against the refuse flow direction, rising together over
the center of the grate into the radiant boiler.
The heat exchange consequences of these three basic furnace designs are derived from the
gas radiation to the bed and the enclosure walls. In this respect, it should be noted that the
radiative balance on the top of the bed is an important boundary condition in setting the
rate of heat transfer down into the refuse bed and in establishing the temperature profile in
the burning mass. Thus, an overbed gas flow that is exceptionally hot or cold has a strong
influence on bed burning.
High heat content, relatively dry refuse does not require much heat return to ignite
and burn completely. The ignition wave moves rapidly into the mass, and heat release
within the bed is more than adequate to sustain combustion. In fact, high heat transfer rates
to the incoming refuse can lead to ignition immediately upon entry (perhaps spreading
back up the feed chute) and=or may push up the bed temperature to levels where ash
softening occurs (with slag buildup on the grate and the side walls). Such increased heat
exchange in the entry area is favored by the contra flow design and may be best for low
heat content, wet refuse. Conversely, parallel flow designs reduce the heat flux at the entry
point; a design that is well suited for dry, high heat content wastes. The center flow system
is a good compromise and is optimal for variable waste.
For almost any initial waste characteristic, high heat transfer is not desired in the ash
discharge zone where ash cooling is desired. Removal of heat from this area is favored by
the contra flow design and, to a degree, by the center flow design.
With parallel flow, the gas flow back into the radiant section impinges on the
front wall such that refractory may be necessary to avoid erosive loss and corrosion
at the impingement point. With contra and center flow designs, the gas flows uniformly
to the radiant furnace and refractory can be minimized (410). The contra flow design
also places the front wall of the radiant boiler close to the partition between the refuse
bunker and the boiler house. This improves the utilization of plan area of the boiler
plant.
In each of these envelope shape concepts, the gases are forced into turbulent
whirling and mixing movements by necking and deflections in the combustion chamber
made by internal noses, baffles, edges, and deflection walls. Mixing is further stimulated
and controlled by the injection of high-velocity secondary air. Water table studies and tests
in full-scale incinerators show that all three enclosure concepts can show acceptable and
essentially equivalent mixing with proper design and secondary air arrangement (410).
However, the parallel flow system shows longer retention time of flue gas flows in the grate
area. Also, the high oxygen demand, complex tarry hydrocarbon gases distilled from the
entering refuse in the contra flow system can rise as CO-rich streaks into the radiant
furnace. To cope with this, mixing must be effected by vigorous secondary air jet attack,
where the mixing effectiveness is often facilitated by constricting the flow field with a
nose (thus increasing the cross-flow velocity differential) at the entry to the radiant
furnace.
It is critical that the furnace gases from the various zones of the grate are provided
with sufficient air to eliminate reducing gases (CO and H2 ) before entering the radiant
furnace. Further, the flow field must not result in impingement or scouring of the waterwall
construction side-walls. Also, the designer must minimize the pickup and carryover of fly
ash. The combination of the exposure of fresh metal surface through erosive processes and

the corrosive effects of reducing gases can be most serious and will lead to unacceptable
rates of metal wastage.
In most cases with variable, midrange heating value waste, the center flow design is
optimal. The contra flow is quite acceptable, even advantageous if the waste is regularly
wet and of low heat content. The parallel flow design should, probably, be reserved for
unusual circumstances where layout, reliably dry, hot refuse and other features of the site
and waste demand its unusual features (410).
3.

Heat Release Rate

The volumetric heat release rate characterizes the combustion intensity and wall temperature level in the furnace enclosure. Although designs vary, most refractory furnaces fall
within the range from 130,000 to 225,000 kcal hr"1 m"3 , with an average of about
180,000 kcal hr"1 m"3 .
Typical heat release rates per meter of furnace width in waterwall boiler systems
approximate 10 million kcal hr"1 m"1 . The primary furnace volumetric heat release rate
approximates 90,000 kcal hr"1 m"3 . This compares unfavorably with stoker-fired coal

Table 2 Heat Release for Various Fuels and Firing Conditions

Stoker

Fuel

Single retort (underfeed)

Bituminous coal
Multiple retort (underfeed) Bituminous coal
Traveling (cross-feed)
Bituminous coal
Municipal refuse
Spreader (overfeed)
on stationary grate
Bituminous coal
Lignite, bagasse
on dumping grate
Bituminous coal
Lignite, bagasse
on oscillating grate
Bituminous coal
Lignite, bagasse
on traveling grate
Bituminous coal
Lignite, bagasse
Pulverized coal:
N=A
dry bottom
N=A
wet bottom
Oil
N=A
normal pressure
N=A
increased pressure
N=A
gas turbine
N=A
Gas
Nuclear reactors
N=A
gas cooled
N=A
Pressurized water reactor
N=A
Boiling water reactor
N=A
Liquid metal cooled

Grate heat
Volumetric
release
heat release
103 kcal hr"1 m"2 106 kcal hr"1 m"3
540
810
810
810

0.260.31
0.18

9501080
12201630
10201290
13601760
1500
17601900
17001970
23002700
N=A
N=A

0.130.20
0.604

N=A
N=A
N=A
N=A
N=A
N=A
N=A
N=A
N=A

0.181
10
100
0.180.25
0.430.86
17160
1726
170

boilers, which operate at 260,000 to 310,000 kcal hr"1 m"3 . Comparative heat release
rates are shown in Table 2.
4.

Secondary Combustion Chambers

Many design specifications require a separate secondary combustion chamber with the
incinerator installation. While the primary combustion chamber may be defined as the
ignition, volatilization, and burning zone above the incinerator grates, the secondary
combustion chamber or zone may be either a separate downstream chamber or an
additional furnace volume downstream of the grate area. In either case, the secondary
combustion zone provides suitable residence time (usually at least 1 to 2 sec) for
completion of the gas phase combustion reactions.
The secondary zone should be located downstream of the point at which additional
air is mixed with the furnace gases to supply oxygen for the completion of the combustion.
The temperature of the secondary combustion chamber should be maintained above 800!
to 900! C to ensure complete oxidation of trace organic pollutants, smoke (carbon and tar
particles), hydrocarbon vapors, and combustible particulate and gases (such as carbon
monoxide). A tertiary chamber is sometimes added downstream of the secondary chamber
in small units to provide additional combustion time, to provide for flue gas cooling and
settling of fly ash, or simply to provide a flue gas connection from the secondary chamber
to the breeching to the air pollution control device and stack.
5.

Slagging and Clinkering

In the design and operation of an incinerator, care should be given to avoiding side-wall,
roof, boiler-tube, or grate temperatures where the refuse ash becomes tacky. Under such
conditions, slag will build on the walls and roof or massive clinkers can form on the refuse
bed. Data or estimates can be made of the temperatures where ash materials become tacky
(the initial deformation temperature in the ASTM ash fusion temperature test). The grate
and the side-walls near the burning refuse bed are likely to operate under reducing
conditions, and the initial deformation temperatures under reducing conditions are the
Table 3 Ash Fusion Temperature Ranges of Refuse and
Coala
Ash source
Refuseb
initial deformation
softening H W)
softening (H 1=2 W)
fluid
Coalc
initial deformation
softening H W
softening H 1=2 W
fluid
a

Reducing
atmosphere
! C

Oxidizing
atmosphere
! C

10301130
12001300
12101310
13151405

11101150
12401325
12551345
13601480

10601100
10801205
11951215
12301425

11051245
11601345
12401355
13101430

From Ref. 130.

From three samples of St. Louis refuse, with magnetic metals removed.
c
From three samples of Union Electric Company coals.
b

proper data to use in assessing the problem. Conversely, the slag screen, upper side-walls,
and roof are generally in a fully oxidizing gas. Table 3 indicates the effect of temperature
and ambient atmosphere on the fusion characteristics of coal and refuse residues.
6.

Temperature and Velocity Distribution

The waterwalls and other heat recovery surfaces define the distribution of gas temperature
and velocity through the boiler. As discussed in this chapter and elsewhere, both
temperature and velocity impact corrosion and erosion losses to boiler tubes. The
design conditions through the boiler vary between manufacturers based on their experience, their selection of materials of construction, their tube layout, and so forth. Typical
actual gas velocities and temperatures are illustrated below:
Actual gas velocities
Primary furnace
Open pass
Convection screen tubes
Superheater
Convection section
Economizer section

m=sec

ft=sec

3.66
6.10
5.49
6.10
6.10
7.62

12
20
18
20
20
25

Maximum temperature entering superheater:

Minimum gas temperature leaving economizer
Desired minimum available air preheat temperature
IX.

650! C
215! C
120! C

ENERGY MARKETS AND ENERGY RECOVERY

As fossil fuel and other energy costs increase, the economic benefits of incorporating
energy recovery into an incineration system improve in kind. However, the analyst should
carefully consider whether the value of recovered energy does exceed the increased capital
and operating expense. And, as important, whether the impact of heat recovery subsystems
on the reliability of the waste disposal system seriously erodes achievement of the primary
function of the incinerator: cost-effective and reliable waste management. For sludge and
solid waste incinerator applications, for example, particular attention should be given to
the areas of corrosion, slagging, and other tube-fouling problems. Generally, the problems
are considerably less important for liquid or fume incinerator boilers.
Energy markets (a useful application or use for recovered heat) are important to
justify incorporating energy recovery into an incinerator design. Energy markets may be
characterized in four ways: (1) the size of the market, (2) the energy type, (3) the reliability
with which energy will be used (the market stability over the year), and (4) the reliability of
revenues.
A.

Market Size

The best markets for waste-derived energy are large markets. Then, fluctuations in
energy recovery from the incinerator (as waste characteristics vary) and=or outright
outages from plant shutdowns can be absorbed by other, on-line energy generators.
Large markets of this type include electric utilities that serve an effectively infinite market

and are well backed up and major steam users (large steam-intensive manufacturing plants
such as paper mills). Smaller users are often concerned about the reliability of energy
supply and may demand backup. Ideally, one wishes to sell an energy product that gives
the customer the opportunity to avoid capital investment (thus adding value beyond energy
cost to the waste-derived energy). If the customer must provide backup steam generation
and a shadow workforce and=or warm-running to allow quick response to outages, this
negates a large fraction of the benefits of waste-to-energy.
B.

Market Type

The ideal market is a steam user. This avoids the need to install energy conversion
equipment (turbo-generators) and the inefficiency in energy conversion for incineration
plants, which often operate with inferior steam conditions (temperature and pressure)
compared to a utility generating plant. Unfortunately, it is often very difficult to find steam
customers using the quantity of steam that fits the incinerator and are located near to the
proposed plant. Also, most steam customers present a risk that, prior to the repayment of
the incinerator bonds, they may go out of business, cut back operations, or significantly
improve energy utilization patterns such that the market (revenue stream) weakens or
disappears.
Electrical generation with sale to a local utility firm feeding the electric grid, has
the significant advantage that the market is essentially infinite in extent and assuredly
continuing. The primary disadvantage is that there exists a highly efficient competitor
(the utility) also generating electricity. Energy credits are often limited to the offset perkilowatt-hour energy cost of fossil fuel. Since the utility price for their fuel is usually low
and their heat rate (fuel energy burned per kilowatt-hour generated) is usually superior to
that of the incinerator, this means the energy revenue can be minimal. Further, the capital
investment in the incinerator energy recovery system may not be recoverable as a
capacity credit associated with the sale of electricity unless the installed utility
generating capacity is marginal (relative to regulatory minimums). Then, the addition of
the incinerator generating capacity to the system has worth that can be realized as part of
the energy credit.
Wheeling of electrical energy (transmission of electricity from the point of generation to a distant customer over the transmission lines of the local, host utility) is, in
concept, a means to sell generated power to a remote utility client that can offer a better
price. This strategy is suggested, for example, when the local utility burns coal and the
remote utility burns oil. Note, however, that wheeling (1) is provided for a fee and (2) is
not a service that a utility must offer under present U.S. energy law. Utilities can (and have)
refused to wheel. Wheeling to a remote commercial client (remote from the point of
generation) is usually prohibited by the exclusivity terms of the utility franchise. In some
circumstances, however, over-the-fence sale to commercial clients (e.g., a large
industrial complex) is feasible.
C.

Market Reliability

The solid waste problem (usually) knows little seasonality. Thus, the ideal energy market is
24 hours per day, 7 days per week, 52 weeks per year. Although this is true for electricity,
most steam markets fall short of this goal. Large steam users such as paper mills come the
closest. Seven-day industrial operations of lesser size may be acceptable for much of the
year. However, white-collar industry, school or office buildings, and other energy

markets primarily using the steam for space conditioning show strong and unfavorable
diurnal and seasonal patterns.
D.

Revenue Reliability

The financial lifetime of incineration systems is long. Therefore, energy marketing

agreements will extend over many cycles of base energy cost (the cost of the reference
fuel often specified in energy contracts and used to scale the unit value of the plants
energy product, such as the cost of No. 2 distillate oil in New York). In some cases,
contract terms at fixed prices may be obtained over relatively long times but usually at
levels that are significantly lower than those that use more risky floating rates. On the
other hand, if the unit value of energy floats, the project revenue stream is uncertain.
This usually requires some kind of backing for the energy-related revenue stream if bonds
(where the payment of principal and interest is based in part on the energy revenues) are
used to finance the project.
Also, as noted above for steam sales, the financial viability of the energy customer
may weaken or fail in the future such that a needed cash flow stream is cut off.
X.

COMBUSTION AIR

For incinerators with grates to support and=or convey the burning refuse through the
furnace, combustion air usually includes the supply of underfire air, overfire air, and
secondary air. Each air supply type serves a special function within the combustion
system.
A.

Underfire Air

While combustion air for hearth-type systems will be discussed later, underfire air for grate
systems is defined as the air supplied under the grate surface through several (typically
three or four) independent plenums to provide a proper distribution of air addition along
the combustion path. Figure 12 shows an example of a 5-plenum underfire air system for a
continuous-feed, reciprocating grate MWC. In the system shown, the plenums are fed from
a common header with dampers to partition the air into each windbox. The air passes
upward through the grates and beneath the burning refuse. When refuse moisture content
is high, the air may be preheated to facilitate drying.
The grate surface meters the primary combustion air, providing a relatively high
pressure drop to ensure good distribution of the air over the grate surface. The air flows
through small air ports or tuye`res in the surface of the individual grate bars with an area
approximately 3% of the total grate area.
The flow of underfire air is required to cool the grates (to maintain their structural
integrity and to avoid oxidative corrosion of the grate metal). Underfire air is also supplied
to furnish oxygen for the gasification and combustion reactions within the bed. In most
regions of the bed, the underfire air only serves to dry, pyrolyze, and=or gasify the refuse
and remove the volatile components. Completion of the combustion reactions (to CO2 and
water vapor) primarily occurs above the bed and involves oxygen from the overfire air.
It is common practice to design for about 70% of the total air to be underfire air and
to operate with about 60% underfire. Thus, the normal underfire air flow rate is slightly
above the theoretical requirement for complete combustion of the refuse on the grates.
However, poor air distribution vis-a-vis air demand, the high relative rate of the gasification
and pyrolysis reactions, and poor gas mixing over the bed require additional air to be

Figure 12 Reciprocating grate MWC showing underfire air supply system. (Courtesy of Von Roll,
Inc.)

supplied as overfire or secondary air for combustion, for mixing, and for tempering of the
gases (dilution) to maintain temperatures below 980! to 1100! C. Limited quantities of
underfire air are required in the last stage of batch incineration process (or near the end of
the grate in the continuous process on grates) for oxidation of the remaining fixed carbon
in the ash residue in order to provide complete burnout.
B.

Overfire Air

Overfire air may be defined as that air admitted above the burning bed of refuse on the
grate. It is usually admitted in high-velocity jets directed such as to thoroughly mix the
combustible gases rising from the burning refuse with combustion air. In some cases (e.g.,
above the dump grates in pulverized coal furnaces burning shredded refuse-derived fuel),
overgrate air is supplied to provide an air curtain at the boiler-tube walls to eliminate the
occurrence of reducing conditions that would foster oxidizingreducing-type tube wastage.
It is common practice to design for about 50% of the total air to be overfire air and to
operate with about 40% underfire.
C.

Secondary Air

Secondary air, added for temperature control, may be admitted through high-velocity jets
in the side walls and roof of the furnace enclosure, either near the upstream end of the

primary furnace or at the transition between the primary and secondary furnaces. Also,
secondary air can be added at low velocity through a slot or small openings in the bridge
wall separating the primary chamber from the secondary chamber. In the latter case,
mixing is dependent more on the shape of the chamber and changes in direction of the
main gas stream than on the energy carried by the air jets. The use of such tempering air
has diminished greatly with the shift from refractory to waterwall incinerator construction.
The cooling effect of the waterwalls is quite sufficient to control gas temperatures, and
secondary air is neither needed nor (due to the high cost of adding to the capacity of APC
equipment) affordable.
D.

Combustion Air Fans

Centrifugal fans with electric motor drives are used to supply combustion air. Since the
fans for high-velocity combustion air must operate at higher tip speeds and therefore with
considerable noise, the fans should be located where their noise will not be a nuisance. In
general, sound-absorbing devices should be considered on these and other high-speed fans.
Low-velocity combustion air can also be admitted through louvers, doors, or other
openings as a result of the negative pressure or draft within the furnace chamber. The
quantity of combustion air admitted through such openings can be controlled by dampers
or by the door opening. This technique, though crude, was useful in small, natural draft
incinerator units. As for other secondary air systems, this approach to incinerator air
supply has disappeared in modern controlled-air and waterwall heat recovery systems.
Combustion air for suspension burning of shredded and air-classified refuse or other
finely divided waste material that does not use grates as the principal site of waste
gasification (even when a burnout grate is provided) requires a different combustion air
concept than that appropriate for a grate system. In suspension burning within a waterwall
furnace or within a refractory furnace, the air that conveys the shredded refuse into the
chamber may be half or all of the theoretical air required for combustion; however, it must
be sufficient to convey and inject the shredded refuse into the furnace. Often, a portion of
the air is supplied through registers that impart some swirl to the injected flow.
Good results have been obtained when air is added tangentially to the furnace
chamber to create a cyclonic action, with the burning mass in the center of the rotating
cyclone and the air injection surrounding the cyclonic flame. If the suspension burning
system includes an auxiliary grate at the bottom of the furnace chamber for completing the
burnout of the ash residue, a small amount of underfire air through this grate and an air
curtain for side-wall protection are desirable. Although high-speed centrifugal fans in
single or multiple stages can be used to supply the air for injection of the shredded refuse
for suspension burning, positive displacement blowers can also be used for injection air as
well as for high-velocity tangential air jets to induce mixing within the cyclone of flame in
the furnace.
E.

Air Preheat

In order to cope with high-moisture wastes, underfire air preheat is used to assist in drying
the waste. Air preheat may be achieved by directly exchanging heat from the incinerator
flue gases to the incoming air or indirectly where the incinerator flue gases generate steam
and the steam is used to heat the combustion air. Mass burn units often have the capability
to provide air preheat. In the Orient, where waste moisture content is often quite high, air

preheat using direct air preheaters to as high as 300! C is practiced. In RDF systems,
preheat from 150! to 175! C is usually available.
1.

Direct Air Heaters

The four main problems encountered with direct air preheaters are
1.

2.
3.
4.

Low-temperature corrosion, if the incoming air is cold enough to cause the

moisture in the flue gas to condense. The HCl and SO2 =SO3 in the flue gases
dissolve in the water to produce strong mineral acids with subsequent corrosion.
High-temperature corrosion, if the incoming flue gases are hot enough to
oxidize the metal in the preheater.
Plugging due to deposition of particulate matter.
Mechanical damage due to stresses induced by large temperature differentials.
These problems are particularly important when the hot gas temperatures exceed
350! to 400! C. Specialized expansion joints and designs to compensate for
differential expansion of the various structures are important solutions to these
problems. Due to the importance of this device to continued operations, extra
care is needed in vendor selection and in the validation of vendor claims.

There are three common designs for air preheaters: countercurrent, concurrent, and
regenerative. In all cases, provision should be made to bleed settled particulate matter
from the hot gas plenums.
a. Countercurrent Flow Systems. Countercurrent tubular air preheaters are common and
consist of a series of tubes fastened into a stationary tube sheet at the top of the unit and a
floating tube sheet at the bottom to allow for expansion. The flue gases pass through the
tubes (which can be easily cleaned by brush) with air passing around the outside of the
tubes. Baffles are arranged to make the preheater multiple pass for the air flow. Although
both horizontal and vertical arrangements are offered, the vertical options are preferable for
the dirty gas found in waste burning. The cold air and the hot flue gases enter at opposite
ends of the unit to maximize the temperature driving force for heat transfer. This reduces
the cost of the unit.
The plate-type unit is constructed of welded envelopes with the air passing through
the inside of the envelopes. Gas sweeps on the outside where the width between envelopes
is large enough to allow easy access for cleaning of ash deposits. Typical clearances are
25 mm to 35 mm.
b. Concurrent Flow Systems. The design of a concurrent flow air preheater is similar to
the countercurrent flow system, but the cold air and hot flue gases enter at the same end.
This has the advantage that the potential for condensation is minimized but at the price of
increased equipment cost due to the lower overall temperature driving force for heat
transfer. Also, the maximum air preheat temperature is limited in the concurrent design.
c. Regenerative Air Preheater. In the regenerative air preheater, heat is transferred from
the flue gas to a heat-retaining mass (often of refractory construction) mounted in a
rotating wheel. The wheel then rotates past a seal to a zone where the combustion air
passes, recovering heat from the heated mass. This type of air preheater is used in some

gas-fired utility boilers but can be problematic if the flue gases contain considerable
particulate matter.
2.

Air to Steam Coil Air Heaters

An alternative means to avoid condensation and achieve a low level of air preheat is to use
a steam coil to warm the combustion air stream. If additional air heating is needed, one can
add a direct, gas-to-gas heat exchanger following the steam coil. This avoids the cost of a
bypass duct and control system in the direct preheater and provides a use for recovered
steam if sale or other beneficial use is not feasible. The primary disadvantage of steam air
heaters arises from the low ultimate air temperature as limited by typical refuse boiler
steam conditions. However, this limitation is not critical when the steam coil is primarily
installed to control condensation. Steam exchangers may be especially economical to
preheat overfire air or secondary chamber combustion air in cases where auxiliary fuel is
required. All the energy in the air preheat directly offsets purchased fuel energy.
Since the undergrate air is commonly drawn from the dusty environment of the pit,
finned tube steam coil air heaters should be designed with generous tube spacing (5 to
6 mm minimum space).

XI.

ASH REMOVAL AND HANDLING

A.

Overview of Ash Problems

Municipal solid waste includes inert materials that cannot be destroyed in the combustion
process. Also, the incineration process is inherently imperfect so that some potentially
combustible material is dried, heated, and carbonized but the desired next step (gasification
of the char) is not achieved. Further, some material simply falls between the cracks
(siftings) and leaves the hot combustion environment substantially unburned. These three
components comprise bottom ash, the inevitable residue of municipal solid waste
incineration operations. Municipal incinerator ash is usually characterized as
Bottom ash: the ash that falls from the grate combined with the siftings that fall
through the grate.
Fly ash: the fine ash that becomes airborne in the primary chamber and either settles
in the ducts and devices of the incinerator or, ultimately, becomes the inlet
loading of particulate matter to the air pollution control system. Also, the fly ash
includes refuse constituents that volatilize in the high-temperature zones of the
furnace and, subsequently, condense on particulate (often the small-diameter
particles that present a large surface area). These may include heavy metals and
high-molecular-weight hydrocarbons with a significant health effect.
The presence of ash imposes several technical and economic stresses on the incineration
operation and the incineration business:
Since ash is a solid and cannot simply be drained from the incineration system,
costly and high-maintenance devices are needed to remove the solids from the
combustor and to handle the ash stream.

Ash (especially the smaller particles in the fly ash) is a concentrate of toxic elements
such as lead, nickel, and mercury as well as elements that are both carcinogenic
and toxic such as cadmium, hexavalent chromium, and arsenic.
Ash constitutes a waste stream of the incinerator, and a place must be found to get
rid of it. This generates a continuing operating cost both for transport to its
disposal site and for the disposal itself. Potentially, landfill disposal leaves the
incinerator firm with a liability for groundwater contamination and other adverse
short- and long-term consequences of residue disposal.
Ash hazards (real or imagined) have emerged in many countries as a significant
concern among the public and the regulatory agencies. These concerns can be
addressed, but they can be an impediment to project implementation.
Ash (especially bottom ash) is not an especially desirable material. It is quite variable
in its properties, including both large clinkers and fine dusts; it may include both
massive and wire metals and ceramic and stony materials; and it exhibits a variety
of colors, mechanical strengths, and other physical and chemical properties.
Other than by the extraction of ferrous metal (easy to accomplish with a simple
magnetic separator), processing the residue to adjust its properties to meet the
demands of the marketplace can be quite costly in comparison to the modest
revenue stream that can be expected.
All of these factors can be important in making the environmental assessments, developing
the operating strategy, and carrying out the economic analysis concerned with municipal
solid waste incineration. Understanding the practical technical alternatives, the legitimate
environmental concerns, and the realistic economic factors associated with ash management is an important element of design, operations, and business planning for refuse
incineration.

B.

Ash Properties

As noted above, ash is variable. Also, the characteristics of incinerator ash are inherently
related to the specific sources and character of the waste fed to the incinerator. The
physical properties of the ash are closely tied to the characteristics of the incineration
process. As a starting point in understanding ash properties, it is useful to review the U.S.and European-based ash data in the tables that follow.
The major elemental analysis of combined bottom and fly ash is shown in Table 4.
The dominant elements are silicon, aluminum, iron, and calcium, much as normal soil
(Table 5). From elements present at the percent level, the material is much like a sandy
or clay-like soil except for the high iron content. Handling and disposal of these
constituents do not present problems. The iron in U.S. and European refuse residue is
primarily present in massive iron and steel shapes (fastenings, iron bars, small appliances,
etc.), which can be easily and usually economically separated from the remainder of the
residue stream.
However, the heavy metal content of the bottom ash and, especially, the fly ash does
present disposal problems to some incinerator operators. Of particular importance has been
concern regarding groundwater contamination. As with the sewage sludge ash case, it has
been observed that the concentration of many of the heavy metals in fly ash is severalfold

Table 4 Range of Major Elements Analysis of Combined Municipal Solid Waste Bottom and Fly
Ash
Element=Compound

Weight percent

Element=Compound

Weight percent

4050
515
0.751.5
1225
815
12
0.751.5

Na2 O
SO3
P2 O5
CuO
PbO
ZnO
LOIa

36
0.51.5
0.50.75
0.060.15
0.040.22
0.120.22
13

SiO2
Al2 O3
TiO2
Fe2 O3
CaO
MgO
K2 O
a

LOI loss on ignition at 750! C.

Source: From (466).

Table 5 Comparison of Ranges of Metal Concentrations in Soil, Refuse Combustibles, Bottom

and Fly Ash (mg=kg)
Metal

Common
soil

Refuse
combustible

Bottom
ash

Fly ash

Al min
max
As min
max
Ca min
max
Cd min
max
Cr min
max
Cu min
max
Fe min
max
Hg min
max
Ni min
max
Pb min
max
Zn min
max

10,000
300,000
1
50
7,000
500,000
0
1
1
1,000
2
100
7,000
550,000
0
0
5
500
2
200
10
300

3,000
25,000
0
15
2,300
50,000
0
90
2
200
20
3,400
500
45,000
0
2
1
90
30
1,600
40
8,000

18,000
177,000
2
2,000
4,100
96,000
0
170
10
2,000
40
18,000
400
480,000
0
4
7
600
30
44,000
90
128,000

31,000
177,000
3
750
33,000
86,000
2
78,000
20
3,000
200
5,000
3,100
320,000
1
100
10
29,000
200
140,000
2,000
280,000

Source: From (323).

the concentration in the bottom ash. Golden et al. (323) drew on a data base of nearly 500
documents in preparing a comparison of the levels of metals in municipal solid waste, and
the bottom ash and fly ash produced. Descriptive statistics for the data (Table 6) were
presented on two bases:

Mean and standard deviation considering all samples as unique no matter how many
samples were derived from a single installation
Mean and standard deviation where averages were first calculated for each
document=source and then the statistics were calculated such that each document=source was regarded as a single data value
The composition data from Table 6 only hint at what has become the defining
characteristic of ash: its content of heavy metals. The term heavy metals, although
commonly used in the literature, is somewhat of a misnomer since several of the metals of
concern are from the light end of the metals spectrum: arsenic, selenium, and beryllium.
In any event, the presence of toxic and carcinogenic metals, even in trace quantities, raises
concern in the regulatory agencies and in the general population and can be a significant
problem in finding secondary uses for residue.
Throughout most of the world, the preferred laboratory methodology with which
to assess the magnitude of the heavy metal problem is the toxicity characteristic
leaching procedure, or TCLP. Under this analysis protocol, prepared samples of the
material under test are mixed with diluted acid and, after holding the mixture at a set
temperature for a specified time, the mixture is filtered and tested for each of a
specified list of elements. Under the U.S. EPA guidelines, the material is said to have
passed the test if the concentrations of the eight specified elements are all less than the
maximum contaminant limit, or MCL. Note that in failing the tests, by U.S. regulation,
the material being tested is designated as a hazardous waste with all of the procedural,
environmental, political, and economic consequences that arise from the assignment of
that characterization.
Table 7 shows TCLP results for combined bottom and fly ash for a Florida
incinerator. The vast majority of tests on U.S. residues do pass the TCLP test. Occasionally
(usually when someone in the waste generation community has dumped an unusually large
amount of a special waste into the refuse stream), the test results indicate that the residue is
hazardous (here, marginally, for lead and cadmium) but, almost invariably, retesting results
in a pass.
Some regulators, upon noting the presence of dioxins, furans, and PCBs in some
samples of ash, have expressed concern regarding proper handling and disposal. Especially
when an electrostatic precipitator (ESP) is used for air pollution control, dioxins and furans
can be created in the ESPs and a fraction of the created material adsorbs onto and is
captured with the particulate matter. However, very strong bonding to solid matter is a
characteristic of all three classes of compounds. Therefore, while the materials can be
found in some ashes, these compounds do not leach to any environmentally significant
degree from the solid matrix.
Table 8 illustrates several features of the ash metal problem.
The range of metal concentration for a given element is large; as much as 100-fold.
The metal content of the various ashes is strongly dependent on the source of the ash. The
finely divided matter (siftings, spray dryer, and baghouse material) is especially
enriched in heavy metal.
Data on the composition of siftings are quite limited. However, the WASTE program
(468) indicated that grate siftings contributed about 3.8% of the total mass flux from a
modern mass burn incinerator. Table 9 illustrates the range of concentration of elements
found in a one-week study of the siftings from the grates of mass burn and RDF municipal
solid waste (MSW) incineration systems.

Table 6 Metal Concentrations in Incinerator Bottom Ash and Fly Ash (323)
Bottom ash (mg=kg)
All samples
Metal
Al
As
Ca
Cd
Cr
Cu
Fe
Hg
Ni
Pb
Zn

Document averages only

No.
doc.

No.
samp.

Mean

Std.
dev.

Mean

Std.
dev.

Maximum

8
7
6
10
11
10
8
8
10
12
13

38
36
23
37
43
42
37
32
41
40
43

32,000
300
50,000
20
3,300
1,700
94,000
1
180
6,000
12,000

17,000
600
28,000
120
23,000
2,600
111,000
2
200
13,000
35,000

31,000
150
49,000
30
2,600
1,700
97,000
1
160
5,400
12,000

13,000
350
13,000
30
7,400
1,100
119,000
1
140
10,000
32,000

177,000
2,000
96,000
70
1,800
18,000
480,000
4
600
44,000
128,000

Fly ash (mg=kg)

All samples
Metal
Al
As
Ca
Cd
Cr
Cu
Fe
Hg
Ni
Pb
Zn

Document averages only

No.
doc.

No.
samp.

Mean

Std.
dev.

Mean

Std.
dev.

Maximum

7
9
6
14
13
11
9
6
11
13
13

57
68
56
84
77
71
68
15
63
80
72

96,000
50
50,000
2,300
1,000
1,000
45,000
20
460
9,300
42,000

67,000
190
31,000
1,000
1,100
900
104,000
40
600
3,400
126,000

80,000
60
57,000
2,000
900
1,000
55,000
20
360
13,000
63,000

45,000
40
26,000
6,000
700
600
78,000
20
500
25,000
97,000

177,000
750
86,000
8,300
2,900
5,000
320,000
100
29,000
140,000
280,000

Combined ash (mg=kg)

All samples
Metal
Al
As
Ca
Cd
Cr
Cu
Fe
Hg
Ni
Pb
Zn

Document averages only

No.
doc.

No.
samp.

Mean

Std.
dev.

Mean

Std.
dev.

Maximum

6
5
4
8
9
7
6
5
8
10
8

48
25
22
38
53
48
44
24
21
64
49

26,000
2
41,000
30
100
3,800
31,000
1
740
1,400
2,600

45,000
6
43,000
100
160
14,000
54,000
2
1,800
2,700
2,700

34,000
3
51,000
60
90
5,700
34,000
4
500
1,900
3,500

28,000
3
31,000
80
80
10,000
29,000
6
1,100
1,500
2,300

106,000
50
67,000
70
1,500
100,000
113,000
18
13,000
8,200
8,000

Table 7 Toxicity Characteristic Leaching Procedure (TCLP)

Results for Combined Municipal Solid Waste Bottom and Fly Ash

Parameter

Leachate (mg=l)

Maximum USEPA
contaminant limit (MCL)a
(mg=l)

Arsenic
Barium
Cadmium
Chromium
Lead
Mercury
Selenium
Silver
Copper
Zinc

<0.002
2.93
1.07
0.024
5.6
0.006
<0.005
<0.01
1.2
120.25

5.0
100.0
1.0
0.5
5.0
0.2
1.0
5.0

Note: 100 times U.S. drinking water standards.

Source: From (466).

Many regulators require the fly-ash dust to be maintained separate from the bottom
ash and mandate periodic TCLP testing. In general, if there is a failure, it is the fly-ash
fraction that fails TCLP testing. If the TCLP criteria are exceeded, the incinerator operator
may be forced to seek a hazardous waste landfill for ash disposal. Due to the inherently
lower metal content and larger mass of the bottom ash, combined ash generally passes the
TCLP.
Because of the significant economic impact of using a hazardous waste landfill for
any or all of the ash, it is highly recommended that the Service Agreement include
language that passes the increased ash disposal cost directly to the community generating
the waste. This is ultimately fair since it is the contamination of the generators waste that
causes the ash to exceed the TCLP criterion. It is not the fault of the operator of the
incinerator. This potential penalty often leads the source community to institute waste rules
such as by citing automobile batteries (a major source of lead) as unacceptable for

Table 8 Range of Total Metal Concentrations in Various Ash Types

Ash type
Bottom Ash
Grate
Siftings
Fly ash
Boiler tube
Spray dryer
Baghouse
Combined
Source: From (465).

No.
samples

Cadmium

Copper

Lead

Zinc

4
60

<1.048.2
<0.6867.6

42012,600
12221,200

3002,750
738103,000

9032,420
41246,100

7
5
56
54

130389
38.059.4
40578
7.7120

534988
312880
1424,399
44517,355

4,28016,100
1,0601,710
1,10010,340
5615,100

11,10024,300
2,8309,630
28092,356
73353,800

Table 9 Range of Concentration of Elements in Grate

Siftings
Concentration range (mg=kg)
Element
Ag
Al
As
Ba
Be
Bi
Ca
Cd
Co
Cr
Cu
Fe
Hg
Mg
Mn
Mo
Na
Ni
P
Pb
Sb
Se
Si
Sn
Te
Ti
V
Zn

Mass burn siftings

RDF grate siftings

38,00063,000
265
1,3492,629
<3.0

<5.014.6

2,34725,215

<0.025.39

169468

5,60034,000
130570
<0.258.6

171946

2,4505,334

<0.9
36,23062,300
813
150990
<0.9
5001,170
51,79092,190
5.312.9
28170
230460
74011,530
25,15036,510
0.22.35
7,66012,780
4901,170
40120
31,80041,600
2101,140
701,560
2,14020,390
659
0.31.8
49,07099,380
4502,000
<2.3
4,3908,990
60170
1,1506,730

Source: From (468).

disposal in the refuse system. Also, such rules encourage the community to institute
programs to recycle and recover mercury from thermometers, hearing aid batteries, and
similar sources.
C.

Bottom Ash

After complete incineration of the refuse, the ash residue drops into an ash chamber or
chute from the end of the grate or kiln. In some instances, a roller is located at the
discharge point on the grate to allow the operator to hold back the residue to allow material
to burn further or when there is a problem with the ash discharger or ash handling
conveyors. Such a roller is illustrated to the right of the last grate element in Fig. 13.
Siftings that have fallen through the grates (which may have been either partially or
completely burned) and collected fly ash also may be conveyed to this ash chamber. The

Figure 13 MARTIN1 discharger. 1, curved-bottom tub; 2, inlet section; 3, outlet chute; 4,

connecting piece; 5, air sealing wall; 6, water level; 7, discharging ram; 8, drop-off edge; 9, driving
shaft; 10, float tank with valve; 11, clinker hopper; and 12, intermediate frame. (Courtesy of Ogden
Energy Group, Inc.)
ash may be discharged directly into a container or onto suitable conveyors for disposal, or
into water for quenching and cooling. The ash residue is then removed from the water with
a hydraulic ram, drag conveyor, pusher conveyor, or other means. Rubber-belt conveyors
have been used in Europe for some time on water-quenched ash residue, and there are now
a few installations in the United States. Alternatively, the ash residue can be handled
manually from the ash hopper by discharging into a dump truck or other suitable container
and transported to the residue disposal site or serviced to by-product recovery systems.
To prevent in-leakage of air (disrupting the combustion air balance in the furnace) or
out-leakage of furnace gases at the point where the gas is removed (impacting air quality in
the working environment), a positive air seal is necessary. Dry mechanical seals and seals
made by covering the ash receptacle or container have been used to control air leakage.
With wet removal of the ash, a wet or hydraulic (water) seal is used or a combination of a
wet and mechanical seal is used. The ash residue usually includes abrasive and corrosive
materials and miscellaneous slag and metal. Therefore, to avoid malfunctions, the effects
of wire, coat hangers, cans, metal hoops, and occasional large and heavy pieces of wood or

metal must be carefully considered in the design of equipment that handles the residue. In
order to keep this critical subsystem of the incinerator functioning, designs must
incorporate features of ruggedness, flexibility, and resilience. Weak, undersized ash
handling systems will cause shutdowns.
Means must be provided for the removal of bottom ash from the primary chamber
without significantly changing the furnace draft or allowing large quantities of tramp air to
enter the combustion chambers. Such upsets in draft or air supply have a major adverse
impact on the combustion process and can lead to overloads in the air flow to the air
pollution control device. Also, large quantities of tramp air cool the furnace gases, perhaps
to temperatures where combustion reactions stop. This increases CO emissions.

1.

Wet Systems

In most plants in the United States, Europe, and Japan, the ash is quenched in a water
trough at the discharge end of the grate. Most plants then use a discharge plunger ram to
push the ash up an inclined ramp, discharging to an apron or vibrating conveyor. An
example of such a ram discharger is shown in Fig. 13. In wet systems, a water seal is
maintained to avoid in-leakage of air. The water quenches any burning material and
minimizes dusting in subsequent ash handling steps. A large portion of the water drains
back from the residue due to compression and the passage up the discharge ramp (the final
ash is about 80% to 85% solids). The ash is then carried out of the incinerator building
using either a belt or vibrating conveyor. Often a steel vibrating conveyor is preferred to
receive the ash as it exits the furnace to minimize wear and cutting problems on a belt
conveyor from the impact of oversize noncombustible matter.
Older U.S. incinerator designs and small modular units quench grate residue in a
trough filled with water (to provide an air seal). The tank is equipped with a drag chain
conveyor to recover the ash. The ash is dragged up an inclined ramp, draining back into the
quench tank. Most often, the chute from the grate discharge to the trough is divided into an
inverted Y leading to two tanks with a pivoted diversion gate so ash can be chuted to
one or the other tank. This duplication is justified by the frequent outages in the ash drag
chain conveyors and the need to keep the incinerator on-line.

2.

Dry Systems

If a dry system is to be used, means are required to ensure that (1) ash is dumped
frequently (avoiding excessive buildup) and (2) door opening actions are properly
sequenced so that only one door is open at any one time. The inner door seals the furnace
against air in-leakage when the outer door is discharging ash directly into a receiving
conveyor or container. An alternative design would use a drag chain or apron conveyor to
receive the ash from the furnace (using the same, two-door design). The conveyor would
discharge into a container.
The space between the doors should be configured as a hopper with sufficient
capacity to hold several times the volume of ash expected between the timed dumping
cycles. The design of the doors, the grate discharge, and the ash holding hopper should be
arranged to ensure that discharging ash or the accumulation of ash in the hopper cannot
block complete closing. Gasketing of the doors should be incorporated into the design to
improve the air seal.

D.

Siftings

Siftings are the fine material that drops through openings in the grate into the air plenums.
Screw conveyors or other appropriate materials handling systems are used to move the
siftings to the bottom ash discharge point. Data on the quantity of siftings are variable as
the quantity generated depends on the characteristics of both the waste and the grate.
E.

Fly Ash

Dry fly-ash handling is usually provided using an enclosed screw conveyor. These
conveyors are low in cost and efficient to handle the fine dusts collected in electrostatic
precipitators and baghouses.
In the simplest configuration, the fly ash is simply combined with the bottom ash.
However, environmental regulatory agencies may require separate disposal of bottom ash
and fly ash. Usually, the bottom ash is regarded as benign material that can be deposited in
any approved landfill while fly ash may require disposal in a monofill (a special,
leachate-controlled cell in a landfill that is used for a particular waste stream).
Some fly ash will accumulate in hoppers beneath boiler passes. Provision must be
made to periodically discharge this material to fly ash or other residue management
system. The material is usually a fine material, smaller than sand. The ash collected in the
hoppers is dry and can be removed through air lock valves. Data on the quantity of heat
recovery system ash are scattered. A study at three German mass burn facilities (469)
provided the data in Table 10. Note that at lower feed rates (lower gas velocities in the
boiler passes) the fraction of fly ash that settles increases.
F.

Materials Recovery from Ash

In a few plants, the bottom ash is processed for ferrous metal recovery. In a few instances,
additional processing of the residue yields materials useful as a fill or for road construction.
1.

Ferrous Metal Recovery

In the United States, Japan and Europe, the quantity of ferrous metal in incinerator residue
ranges from 6% to 9% by weight. At this concentration, if one takes credit for the avoided
cost of not having to landfill the ferrous material and with typical low-grade scrap metal
prices, recovery of the ferrous metal is, at best, only marginally justified.
The technology of ferrous metal recovery is simple, the capital and operating cost, is
low, and the installation has little impact on plant layout or staffing requirements. However,
the uncertainty of the ferrous scrap market strongly indicates that no firm revenue stream
Table 10 Fly-Ash Deposition in the Flues of Mass Burn Boilers
Facility

Waste feed rate

(mg=hr)

Fly-ash loading
(kg=mg of waste)

Percent captured in
boiler

12
10
10
10

20
20
30
5

7.4
12.5
11
27

A
A
B
C
Source: From (469).

should be assigned to ferrous recovery. Instead, it is usually prudent to consider ferrous

recovery as a cost center, with revenue, if it comes, a welcome benefit. Some client
communities desire the environmental benefits of materials recycling and may specify
ferrous recovery. In such cases, often the service agreement can be structured as a pass
through so that the community pays the incremental costs of the ferrous recovery
operation and is assigned all (or a portion) of the revenue stream generated from sales
of scrap.
Using ballistic and magnetic separation equipment, ferrous metal can be extracted
from the overall residue stream for salvage as scrap. Ferrous recovery is generally effected
by use of a belt electromagnet. Most often, the magnetic belt is located at a transfer point
for the residue conveyor and the recovered ferrous is cast into a chute to a second receiving
container. Often, two stages of magnetic separation are required to produce a saleable
ferrous product from municipal residue. This may be accomplished using multiple drums
or, for belt-type systems, with multiple magnets.
2.

Roadbeds and Earthworks

Following processing for ferrous metal removal, the medium ash solids, such as clinker
particles, portions of fused glass, or particles of shattered glass that pass the magnet, can
sometimes be used as fill material or in surfacing and construction of alleys and secondary
streets.
The oversize noncombustible in the ash residue from the incinerator furnace is
customarily transported directly to the residue disposal site. In a few instances, oversize
noncombustible is passed through crushing or breaker rolls, which flatten larger metal
pieces and crush nonmetal pieces.
Incinerator bottom ash can be used in roadbeds and in earthworks (as clean fill") if
it meets applicable environmental quality and use guidelines. The environmental requirements typical in Europe are
Free of fly ash.
Stored for at least 6 weeks prior to application.
Toxicity characteristic leaching procedure or (TCLP) results must comply with
standards.
Loss on ignition (LOI) less than 6%.
Putrescible material less than 2%.
Typical use constraints include
Emplacement must be at least 0.5 m above the maximum water table.
Use in soil or groundwater preservation areas is restricted.
Measures should be taken to minimize leaching by rainwater (e.g., by use as an
underlayment to an asphalt roadway).
A side and top layer of impervious material must be applied within 6 weeks of
emplacement of the ash roadbed.
The use should focus on larger projects, and the location of the emplacement should
be noted in the as built final drawings.
To improve the acceptance of the ash as a building material, the size distribution must
mesh with the expectations and needs of the user. In the Netherlands, the size distribution
shown in Table 11 meets the size requirements. To meet these size requirements, crushing
and sieving and magnetic removal of ferrous contaminants are necessary processing steps.

Table 11 Granulometry of Certified Municipal Solid Waste Bottom Ash

Coarse type
on sieve
45 mm
22.4 mm
8 mm
2 mm
63 mm

Roadbase material
(0=20) unbonded
% (m=m)

Roadbase material
(0=20) cement bonded
% (m=m)

Embankment material
(0=40)

0
010
1040
4070
92100

nil
010
1040
4070
nil

010
nil
nil
nil
92100

Source: From (463).

The fine ash that has no salvage value is usually transported by truck to a residue
disposal site. Because the fine ash is often enriched with heavy metals, these residues are
usually segregated within the landfill in a separate cell (a monofill"). This minimizes the
potential leaching of metals occasioned by percolation through the residue mass of organic
acids formed as a normal product of the biological breakdown of refuse within the landfill.
3.

Vitrification

The several tables presented above indicate that incinerator residue contains relatively high
concentrations of heavy metals and that the leaching of these metals from the residue is,
for some regulatory agencies, a matter of concern. Over long times, leaching will occur
and, if the residue is not placed in landfills with adequate leachate control and treatment
systems, there is a potential for groundwater contamination. In the United States and in
some other countries, the legal liability of the groundwater contamination resides with the
waste generator (here, the incinerator operator) even though many years may have passed.
All of these problems and the general problem of finding acceptable disposal sites for
residue have led some jurisdictions to vitrify (melt) the residue to a glass. In the vitrifier,
the residue is heated to the fusion point whereupon metals are bound into the matrix of the
glass and rendered substantially insoluble. Residue processing by vitrification allows use
of the residue as an aggregate or as a clean fill without concern regarding environmental
contamination impacts.
Vitrification technology has been studied in several countries but is primarily
practiced in Japan [see (467) and the Kubota furnace described in Section 3.A in Chapter
9]. In Japan, environmental concerns coupled with extreme cost and siting difficulties for
landfills encourage consideration and, in some cases, implementation of vitrification
schemes. Vitrification technologies involve use of electric arc furnaces, high-temperature
combustion, and other means to raise the residue temperature to above 1500! C. In some
instances, lime is added to modify the melting temperature. Under most circumstances, the
cost for vitrification equipment and the expense of purchased energy to effect vitrification
far exceed the cost of landfill disposal options.
4.

Financial Considerations in Residue Processing

Residue is usually not regarded as an incineration by-product of great quality or worth. Its
variable properties generally make its use in high-specification products impossible. In
most cases, the residue must be transported (another cost) to another location for use.

Although residue represents a large problem to the incinerator, compared to competitive

aggregate or fill materials (sand, gravel, etc.), the quantities of residue generated are very
small compared with the market. In many cases, crushing and sieving are needed to
produce a salable, specification raw material. This further increases the price necessary for
a break-even operation. Finally, the cost of sales (sales staff, advertising, applications
engineering, etc.) is large in comparison to the quantity of material generated. All of these
factors mitigate against the ability to turn residue into a money-making resource. Also, ash
processing is a distraction to the principal function of the incinerator and is an improbable
and unreliable revenue source.

XII.

FLUE GAS CONDITIONING

Flue gas conditioning is defined as the cooling of the flue gas after it has left the
combustion zone to permit discharge to mechanical equipment such as dry air pollution
control devices and fans or a stack. Conditioning can be effected by water evaporation as
an alternative to heat removal in a boiler-type incinerator. Water evaporation conditioning
can also be installed as an add-on to boiler systems to lower gas temperatures below those
considered safe in a boiler (due to acid gas condensation) in order to improve the removal
efficiency of acid gases, mercury, dioxin=furan compounds, and some other pollutants. In
general, cooling to 230! to 370! C is necessary if the gas is discharged to mechanical
equipment, while cooling to 470! to 590! C is adequate for discharge to a refractory-lined
stack.
Both wet and dry methods are used for cooling (or tempering) incinerator flue gas
streams. The sections below discuss the technical and economic features of several
embodiments of these methods.
A.

Cooling by Water Evaporation

In wet methods, water is introduced into the hot gas stream and evaporation occurs. The
degree of cooling is controlled by (1) the amount and droplet size of the water added to the
gas and (2) the residence time of the gas in the water atmosphere. The quantity of water
needed to effect a given reduction in gas temperature is readily calculated using a heat
balance on the gas and cooling water as shown in Example 1.
Determine the amount of water (available at 30! C) needed to cool a
hot air stream to 150! C from an initial temperature of 1000! C. Determine the percentage
volume change in comparison to the use of a boiler to achieve the same result. Assume
operation at a pressure of 1 atm.
EXAMPLE 1.

Based on the heat capacity correlations given in Section I of Chapter 2, calculate the
enthalpy per one mol of air between the two temperatures. For simplicity, use the
average molal heat capacity correlations from Fig. 1 of Chapter 2.
H150 6:95150 " 15 938 kcal=kg mol
H1000 7:551000 " 15 7437 kcal kg mol
enthalpy change 6499 kcal=kg mol
From steam tables or from the water=steam enthalpy values available from the
computer program provided with this book, find the water enthalpy (kcal=kg) at

the two temperatures (liquid water supplied at 30! C and steam leaving the cooler
at 150! C).
H30 30:0 kcal=kg
H150 663:2 kcal=kg
enthalpy change 633:2 kcal=kg
Calculate the water required.
water required 6499=633:2 10:26 kg H2 O per mol of air
0:57 kg mols H2 O
Evaluate the volume change.
With a boiler, the volume change of the air is easily determined by the ratio of the
absolute temperatures at inlet and outlet:
volume ratio

150 273:15'
33:26%
1000 273:15'

With water evaporation:

initial volume nRT =P 1:00:082061000 273:15=1:0 104:47 m3
final volume nRT =P 1:570:08206150 273:15=1:0 54:52 m3
volume ratio 54:52100=104:47 52:18%
Presently, two types of wet cooling are used: the wet bottom method and the dry bottom
method.
1.

Wet Bottom Methods

The wet bottom method was the most common method used in the 1960s and 1970s but
has become unpopular since then due to the costs and problems of the associated
wastewater disposal. This approach involved the flow of large quantities of water (much
more than was required for cooling the flue gas). The water was supplied by coarse sprays
operated at relatively low pressures. The excess water fell to the bottom of the cooling zone
and was rejected or recycled. Although not efficient, some removal of particulate occurred
by droplet impingement and gravity settling in the spray chamber.
The equipment used in this system consists of several banks of sprays, each with
several nozzles with relatively large openings (over 0.5 cm). These spray banks are
normally located in the flue leading to the stack or air pollution control equipment.
Pumps are not normally required, since line pressure is adequate for satisfactory operation.
However, pumps are required if the water is recycled. The bottom of the flue must be
maintained watertight to prevent leakage.
The system is generally controlled by measuring the gas temperature downstream of
the sprays and modulating water flow, either manually or automatically. Generally, spare
spray banks are provided that can be pressed into service quickly in response to sudden
temperature increases or plugging of the spray nozzles.
The advantages of this system are that it is relatively simple, reliable, and
inexpensive to design and install. The gas is humidified during cooling (approaching
saturation conditions), which may be somewhat advantageous for some types of air
pollution control equipment (electrostatic precipitators) and disadvantageous for others

(fabric filters). Also, there is a reduction in total gas volume during cooling, as shown in
Fig. 14. This figure shows the change in volume of flue gas as a function of inlet gas
temperature when the gas is being cooled to 260! C by means of air dilution, water
evaporation, and a boiler.
A disadvantage of the wet bottom system is that much more water (greater than
100% excess) is used than is necessary for cooling the gas. The excess water is acidified in
use and is contaminated with particulate and dissolved solids. Therefore, it normally
requires treatment before it can be discharged from the plant. In some installations the
excess water is recycled to the process, and in these cases severe problems of erosion,
corrosion, and plugging are encountered in the nozzles, piping, and recirculation pumps.
As a result, excessive maintenance costs and equipment down-time are common.
Removing suspended solids using clarifiers, hydraulic cyclones, and settling ponds is
helpful, but problems with corrosion, plugging, and erosion have not been eliminated. In
some plants, the clarifiers and ponds are sources of odor. Also, ponds require considerable
land area and may result in groundwater pollution.
Corrosion problems result from the very low pH (less than 2.0) that occurs when
recirculation systems are used without alkali addition. There is also a significant
accumulation of slightly soluble salts in the recirculated water. These problems can be
controlled to some extent by adding chemicals such as lime or soda ash to the
contaminated water. However, such systems have significant operating costs and do not
completely alleviate corrosion and scale problems.
Another disadvantage of the wet bottom method is that flue gas leaving the spray
chamber may carry entrained water droplets or wet particulate matter. These moist
particles can cause operating problems with the APC and fan equipment due to fly-ash
adherence and accumulation. Also, corrosion of unprotected metal parts downstream or
buildup of hardened deposits of solids in the flues and fan housing can occur.

2.

Dry Bottom Methods

In the dry bottom method, only enough water is added to cool the gas to a preset
temperature and the system is designed and operated to ensure complete evaporation. In
this system, a conditioning tower 10 to 30 m high is required and fine, high pressure spray
nozzles are used. Booster pumps are necessary to raise the water pressure to assure fine
atomization; water pressures from 6 to 36 atm are common in such systems. Alternatively,
atomization is effected using compressed air or steam or high-speed rotary atomizers.
Control is usually accomplished with a temperature controller measuring the outlet flue gas
temperature and modulating the flow of water to the conditioning tower sprays. Since high
pressures are required, only a small degree of flow rate modulation by pressure variation is
practical. If the flow is too low, the pressure drop across the nozzles will fall, and poor
atomization will result; if the flow is too great, the pressure drop across the nozzle will
increase and limit flow. As a result, additional spray heads are put into or taken out of
service as required to maintain uniform outlet temperatures. Air atomization techniques are
preferred to obtain the required turn-down.
Water droplet size is very important in the design of a dry bottom conditioning
tower; droplets that are too large will not completely evaporate and, in addition to the
problems of water carryover described above, will cause a wet bottom in the conditioning

Figure 14 Change in flue-gas volume during cooling to 260! C (to 315! C for boiler). [From (21)].
tower. Droplets that are too fine will lead to high power usage. A balance must be made
among droplet size, gas residence time in the tower, and power usage. Equation (1)
presents a formula developed by Hardison, of U.O.P. Air Correction Division (131), for
estimating the evaporation time of water droplets:
t

rd
0:123 T " Td

where
t residence time in sec
rd droplet radius in microns
T temperature of gas ! C
Td temperature of droplet ! C
A second and more complex method of estimating (in a step-wise fashion) the evaporation
time of droplets (up to about 600 mm uses the graph shown in Fig. 15, where
Ta gas temperature K
DT Ta " 54 ! C
m gas viscosity at Ta g cm"1 sec"1
b 0:071 m2 Ta "0:36
If the drop temperature varies significantly from 54! C, the evaporation time may be
corrected by a factor (Ta " 54=Ta " Tdrop . It is suggested that users of this latter method
consult the original paper (132).
Although both of these methods are useful in estimating the residence time required
for evaporation, care must be taken in selecting a nozzle that provides uniform-size
droplets rather than droplets with a wide size distribution, since the largest droplet will
dictate the length of the conditioning tower.

Figure 15 Theoretical evaporation time for water droplets in hot gas streams. [From (132)].

The advantages of the dry bottom system are that it minimizes water consumption,
eliminates water pollution problems, produces a dry effluent gas (free of entrained
water), yet significantly reduces the volume of the flue gas (Fig. 14). The gas is humidified
during cooling, which may or may not be desirable. The disadvantages of the system are
that it is expensive to design and install, power consumption is high, control is somewhat
complex, and the small orifices of the atomizing nozzles make them susceptible to
plugging. The nozzle plugging can be minimized by filtration of the water, proper draining
of the nozzles when they are not in use, and maintenance of sufficient flow through the
nozzles to keep them cool. Generally, the preferred alternative is to use air, steam, or
mechanical atomization methods such as those used in the spray dryer absorber acid gas
control technology.
An installation in New York City used a spray head in which a portion of the water
flow was recycled, thus providing additional nozzle cooling. Keeping the nozzles cool and
draining them are essential to prevent scale formation in the nozzles due to water hardness;
scale is the principal cause of plugging. In order to minimize the risk of outages, the
nozzles should be easily accessible and removable during operations, and additional
nozzles should be provided for emergency use.
The dry bottom system is more costly than the wet bottom system because of the
need for water filtration and the maintenance required for the atomization, pumping,
nozzles, and control systems. Proper materials of construction (such as ceramic wear
surfaces) are essential for good nozzle life and minimizing corrosion. Power consumption
can be minimized by proper nozzle selection.
B.

Cooling by Air Dilution

Dilution with air is the simplest and most reliable method for flue gas cooling. Only a
damper for air control is necessary for a system with adequate draft. Although air dilution
has these advantages, large quantities of air are required for air dilution cooling, increasing
the demands placed on the air pollution control equipment and the induced draft fan and
stack. The increase in volume caused by air cooling is shown in Fig. 14. Such large
increases in volume significantly increase the capital and operating costs of the equipment
that follows the point at which dilution takes place. For these reasons, air dilution is no
longer used in new incinerator installations.
C.

Cooling by Heat Withdrawal

The third method of dry conditioning uses a convection boiler in which heat is removed
from the flue gas by the generation of steam. The equipment consists of a convection tube
waste heat boiler, an economizer, and all of the auxiliary equipment required, such as
boiler feed water pumps, steam drums, and water treatment facilities. In addition, an air or
water condenser (the latter served by a nearby natural cooling water source or cooling
tower) may be necessary to condense the steam when the steam demand is less than the
generation rate. These boilers would have controls similar to conventional boilers and
could require an additional full-time operatora licensed stationary engineer.
The advantages of this system are that heat is recovered and the shrinkage in flue gas
is greater than with any other method discussed, as Fig. 14 shows. No water is added to the
system during cooling, which may or may not be a benefit. In some cases, the steam
generated in the boiler can be used for space heating or process applications (offsetting the
burning of fossil fuel) or can be sold to produce income. Alternatively, steam can be used

to generate electricity for internal use or sale. Finally, steam can, in concept, be used in the
plant to induce mixing in the incinerator furnace gases (steam jets).
Heat recovery (with 100% energy rejection to an air condenser) followed by a dry
baghouse APC system has been found attractive for transportable, short-operating-life
hazardous waste incineration systems where the freedom from concern regarding water
supply for scrubbers and for wastewater disposal may be an important benefit when
coupled with the high collection efficiency of the baghouse.
The disadvantages of heat recovery methods (assuming no attractive energy market
exists within economically attractive distances) are that the system is expensive to design
and install, the boiler installation is complex to operate and may require an experienced
licensed operator, and corrosion and=or erosion problems with the boiler tubes (leading to
system outages) may arise if the design is incorrect.
D.

Steam Plumes

In the incineration process, water is introduced into the flue gas from the evaporation of
refuse moisture and as a result of the combustion process. Water may also be added during
conditioning of the gases and by wet scrubbing. As a result, the flue gas leaving the stack
can contain significant amounts of water vapor, which will condense, under certain
atmospheric conditions, and an opaque stack effluent (steam plume) will result.
Although steam plumes neither cause nor indicate air pollution (often quite the contrary),
public reaction is frequently negative and vocal. While these plumes will appear white and
not unpleasing in bright sunlight, on cloudy days and at twilight, they may seem dark and
dirty. Under any circumstances, they prominently announce to the neighbors that the
incinerator is operating.
Formation of a steam plume depends on the discharge temperature and moisture
content of the flue gas, the ambient temperature, and the amount of dilution that occurs at
the point of flue gas discharge. The processes that occur upon discharge of the stack gases
into the ambient are as follows:
The gases rise from the stack following the flow pattern of a free turbulent
nonisothermal jet (sometimes in crossflow due to winds). One characteristic of
this flow pattern is that little mixing of the jet fluid with the ambient occurs for a
distance of about one stack diameter. As a result, the plume from superheated
vapors may float above the top of the stack.
The stack gases then begin to mix with cooler ambient air. As this occurs, the
temperature and absolute humidity of the mixed gas fall, about in proportion to
the dilution ratio (kg of stack gas per kg of ambient fluid). Thus, the state of the
gas mixture, as shown on a temperaturehumidity (psychrometric) chart, moves
downward and to the left (see Fig. 16). The curve describing the gas state differs
from a straight line (between the stack discharge and ambient temperature and
humidity point) only as a result of differences in relative specific heat between the
flue gas and the ambient air. This effect is often neglected, however, as it acts, in
the analysis which follows, to yield a conservative estimate.
The mixed-gas state thus moves from stack conditions toward cooler and less moist
conditions. If, in the course of this change, the mixed-gas conditions cross the
saturation curve , a steam plume will form. The gaseous part of the plume then
moves down the saturation curve while liquid water, in the form of small

Figure 16 Psychometric chart illustrating steam plume formation. [From (21))].

droplets, is released. The resultant plume opacity is related to the amount of water
condensed, the dilution process (which affects cooling rate and droplet number
and size distribution), and the ambient temperature and humidity.
As dilution proceeds, the condition of the gaseous part of the plume drops from the
saturation curve, and the liquid droplets evaporate. At this point, the plume has
dissipated, and any residual opacity is due to suspended particulate or aerosols.
Typically, however, the high dilution at this distance from the stack effectively
hides what may have been an intolerable plume opacity caused by high fly-ash
loading.
In the case where typical municipal incinerator flue gas is evaporatively cooled from 900 to
260! C, the treated gas will contain approximately 0.36 kg of water vapor per kg of dry air.
This moisture includes water formed by refuse combustion in addition to that evaporated
in cooling the flue gases. Under these conditions, a line drawn on a psychrometric chart
(see Fig. 16) starting at 260! C, 0.36 absolute humidity, and tangent to the saturation line
delineates the set of atmospheric conditions under which a steam plume will form. The
shaded area on Fig. 16 shows these conditions. For this flue gas, a plume will form any
time the ambient temperature falls below 4! C. The plume will also form at higher ambient
temperatures, depending on the absolute humidity of the ambient (falling within the
shaded area).
If the flue gas is cooled to lower temperatures, say to 90! to 105! C in a wet scrubber,
or if the ambient relative humidity is high, plumes will form at higher ambient
temperatures. High cross-winds, providing more rapid dilution, may shorten or almost
eliminate the plume. Reheating of flue gas, though costly, will reduce the intensity or
frequency of plume formation.
Since the costs of reheating will be excessive in regions of frequent low ambient
temperature, it is likely that steam plumes (though possibly seasonal) will always be
associated with incinerators equipped with wet scrubbers and with plants using evaporative
cooling for flue gas conditioning. Steam plumes may be reduced by cooling the flue gases
with a secondary scrubber (subcooler) to condense a portion of the water vapor before
discharge. A modicum of reheating is still necessary to totally avoid a plume. The heat
removal required to condense this water must be compensated for with a cooling tower
system or by using very large quantities of subcooling water. With the cooling tower
approach, one has only moved the steam plume from the stack to the adjacent cooling
tower. The strategy of using very large quantities of water is an approach generally limited
to wastewater treatment plants where the water supply (effluent) is abundant and cheap.
References (133) and (134) further discuss the steam plume problem.

XIII.

ENVIRONMENTAL POLLUTION CONTROL

An incinerator is probably of greatest concern to a municipality because of the fear of the

air pollution impact on the contiguous environment. Water pollution from the residue is a
secondary issue. However, an incinerator can also create undesirable noise and cause the
surrounding area to be unattractive because of litter and other forms of trash that quickly
disfigure an incinerator site where good housekeeping is not regarded as a fundamental
plant responsibility.

A.

Air Pollution

The most noticeable forms of air pollution are fly ash, smoke, odors (from the stack as well
as other areas), noxious gases, and dust. All emanate from an incinerator at times.
1.

Composition of the Flue Gases

If combustion of the volatile fraction of the refuse is complete, the composition of the flue
gas will be principally nitrogen, oxygen, water vapor, and carbon dioxide. There will be
small amounts of sulfur oxides, nitrogen oxides, and mineral acids (principally hydrochloric acid, which will result from the combustion of halogenated plastics, particularly
polyvinyl chloride). Normally, the concentration of sulfur oxides, nitrogen oxides, and
mineral acids will be high enough so that it will trigger the regulatory requirement for air
pollution control. If combustion of the volatile is not complete, the flue gases will contain
significant amounts of carbon monoxide and other unburned or partly burned organic
materials. These emissions are more subtle but can include the polychlorinated dibenzo
p-dioxin and dibenzo furan compounds, POMs, and the like. The first easily visible
indication of the presence of these materials in high concentrations will be the appearance
of black smoke from the incinerator stack, which may be followed by the detection of
objectionable odors.
The presence of unburned or partially burned materials is unnecessary and is caused
by the poor operation of the incinerator. Their emissions can, in general, be controlled by
the proper operation of the incinerator rather than the installation of control devices.
Complete combustion can be ensured by operating the incinerator at the proper temperatures (from 750! to 1000! C), by providing sufficient air for combustion, by providing
sufficient residence time for the combustion process to occur, and by inducing (by both gas
passage configuration and well-designed overfire air jets) sufficient turbulence in the
combustion space to mix the combustible gases and aerosols with the necessary air.
Such residence time and some mixing are usually provided for by ducting the flue
gases to a secondary combustion chamber or zone. Although it is not essential that such a
chamber be provided, it is necessary to provide sufficient volume preceded by vigorous
induced mixing in the furnace to ensure that the combustion process is completed. Few
single-chamber incinerators meet this need.
2.

Particulate Matter

Particulate matter (characterized by flue gas weight loading), generally referred to as fly
ash, is generated in and elutriated by the combustion process and must be removed from
the effluent gases. The amount of particulate matter generated is somewhat dependent on
the design and operation of the incinerator. If the combustion process is not complete, a
sooty fly ash will result. The best way to control emissions of carbonaceous particulate is
operation at temperatures sufficiently high to ensure complete combustion of these
materials.
Studies have been made (21) that indicate a correlation between the amount of fly
ash entrained in the effluent gases and the distribution and amount of overfire and underfire
air and the type of grate employed. Proper operation will ensure that large amounts of fly
ash do not become entrained in the gas stream because of improper air distribution.
No matter how carefully the incinerator is operated, however, particulate matter will
be entrained in the effluent gases. The extent to which the particulate matter is removed
from the gases depends on the type of emission control equipment used and the way it is

operated and maintained. If abnormal amounts of particulate are being emitted, it may be
that the incinerator is being operated improperly. This happens, for example, when
combustion is quenched by large amounts of air admitted to the incinerator in an
uncontrolled manner, such as occurs in batch-feed incinerators. Also, the emission control
equipment may not be operating properly. Problems that could occur in the operation of air
pollution control equipment include the plugging of spray nozzles caused by either water
hardness or large amounts of solid in recirculated liquid streams, the plugging of cyclones
caused by sticky particulate or condensation, failures in the electrical systems of a
precipitator, and fabric filter bag failures due to flexing or pinholing from burns (especially
from the carryover of still-burning sparklers).
In a properly designed and operated incinerator, equipped with appropriate air
pollution control equipment, the stringent standards established by states and the federal
government can be met.
Although the flue gases from incinerators contain a number of pollutants, air
pollution control equipment installed on these units are primarily directed at the problem
of particulate removal. For this purpose, a number of devices are in use, ranging in
particulate removal efficiency from 5% to 15% to above 99%. In light of present and
forecast particulate emission standards throughout the world, control efficiencies in excess
of 98% are generally required.
The electrostatic precipitator (ESP) and fabric filter are currently receiving the
greatest attention for particulate removal from incinerator flue gases. Both systems can be
used in conjunction with the spray dryer absorber (fed with lime slurry) for acid gas
control. Trends at this writing favor the fabric filter since it has a higher fine-particle
control efficiency (important in metal control) and it offers the advantage that the captured
filter cake gives a second contact opportunity for acid gas control vis-a-vis the ESP when
used with a spray dryer absorber. A summary of expected average control efficiencies of
various air pollution control systems is given in Table 12.
3.

Smoke

Smoke (characterized by flue gas opacity) may be ash material that is completely burned
but is very finely divided. More commonly, smoke is predominantly unburned or partly
burned combustible material (tar or soot aerosol). If the smoke is an ash material, control
requires the use of very high efficiency emission control equipment. It may be lessened to
some extent through improved operation by adjusting the air distribution in the primary
furnace.
If the smoke is combustible, it can best be controlled by improved combustion
efficiency. Longer residence time, better air distribution, gas mixing, and high operating
temperatures will eliminate such pollution.
4.

Odor

In addition to odors that emanate from the stack because of poor combustion, there are
odors in the pit and in the residue handling area, particularly if water is used for quenching
and sluicing the grate siftings. Odors may also emanate from the residue disposal landfill.
The odors from the stack can be best controlled by ensuring complete combustion of
the gases. This is done by proper air distribution, elevated temperatures, and proper
residence time at these temperatures. Such control is much more effective and less costly
than using equipment such as scrubbers for odor removal.

20
20
33
40
35
50
70
90
99
99.9
N=A
N=A
N=A

Mineral
particulate
2
2
4
5
7
10
30
80
90
99
N=A
N=A
N=A

Combustible
particulateb

From (26).
Assumed <5 microns.
c
Assumed 2=3 condensed on particulate, 1=3 as vapor.
d
Assumed <5 microns.
e
Some removal occurs for higher molecular weight compounds.

None (flue setting only)a

Dry expansion chambera
Wet bottom expansion chambera
Spray chambera
Wetted wall chambera
Wetted, close-spaced bafflesa
Mechanical cyclones (dry)a
Medium-energy scrubber (wet)a
Electrostatic precipitatora
Fabric filtera
Carbon adsorption system
Dry absorption system
Wetdry absorption system

APC technology
0
0
0
0
0
0
0
0
0
0
N=A
0
0

Carbon
monoxide
0
0
7
25
25
30
0
65
0
0
N=A
0
0

Nitrogen
oxides
0
0
0
0
0
0
0
0
0
0
0e
0
0

Hydrocarbons
0
0
0
10
10
50
0
15
0
0
N=A
75
85

Sulfur
oxides
0
0
10
40
40
50
0
95
0
0
N=A
80
95

Hydrogen
chloride

APC system removal efficiency (weight percent)

Table 12 Average Control Efficiency of Air Pollution Control (APC) Systems

10
10
22
40
40
85
35
95
60
67

Polynuclear
hydrocarbonsc

2
0
4
5
7
10
0
80
90
99

Volatile
metalsd

Odors in the pit can be controlled by using masking deodorants, minimizing the time
the refuse is stored in the pit, planning the withdrawal of material from the pit (a first-in,
first-out strategy), and periodic clean-out of the pit contents. The deodorants can be
applied by using a spray network permanently installed in the pit. However, such a system
should be used only in conjunction with a good cleaning program. The crane operator
should establish a systematic plan for withdrawing refuse from the pit to ensure that no
refuse stays in the pit too long (as by working from one end of the pit to the other). Cleaning
should be done once a week by washing the pit down. This can be done without completely
emptying the pit by scheduling the washing of only portions of the pit at one time.
Odor control in water systems used for quenching the ash and sluicing the grate
siftings can only be accomplished by a very active cleaning program that involves
changing the water at least once a week and throughly washing the equipment. Wet ash
handling systems are, at best, messy and difficult to keep clean, particularly if burnout is
not complete.
5.

Dust

Dust is emitted from the pit area when trucks unload and when the crane is operating, from
the residue handling area, and from the fly-ash handling area. Dust raised during unloading
can be controlled by enclosing the tipping area and by exhausting the tipping floor volume
to the combustion air fan intakes. Also, sprays can be used to moisten the refuse in the pit
to minimize dust during feeding. Dust in the ash handling area can be controlled by using
sprays to moisten the ash and using closed transfer and hopper systems to minimize dust
leakage. Some systems of these types are currently being used and have been very
effective.
B.

Water Pollution

Water is used and immediately discharged from most incinerators. Water is used to wash
down the pit and ash handling systems, to quench hot residue and fly ash, to control dust,
and to aid in gas quenching and scrubbing. Water rejected from the incinerator will most
likely be polluted in one way or another.
The wash water used throughout the plant and the water that drains from the pit and
ash handling systems will have high biological and chemical oxygen demands and contain
suspended and dissolved solids, both organic and inorganic (135, 136). Since the quantity
of this water will be relatively small, it may be discharged into a sanitary sewer or septic
system. It should not be discharged to storm drains or streams without treatment.
The water rejected from quenching and scrubbing systems may be quite large in
quantity and very hot. It will also be high in suspended and dissolved solids and, unless
treated with alkali, will have a low pH. The solids will be both organic and inorganic. The
biological and chemical oxygen demand of these waters may also be higher than desirable.
In some designs, the scrubbing water is settled to remove suspended solids and recycled; in
others, it is discharged after use. Recycling has often proved to be troublesome because of
the plugging of lines and spray nozzles with solids and scale.
Discharged water should be sent to a well-designed settling pond to remove solids
and permit cooling. If the water is to be discharged to a stream, secondary treatment
consisting of pH adjustment and possibly digestion to reduce biological and chemical
oxygen demand may be required. The rate of discharge may have to be controlled to
prevent the buildup of soluble solids in the stream.

If water is used only for quenching, systems can often be designed to obtain 100%
evaporation and thereby eliminate any disposal problem. Wet scrubbers that permit
recirculation without plugging are also highly desirable.
C.

Noise Pollution

The noise created by the operation of an incinerator can be a disagreeable nuisance. Many
large trucks are constantly delivering and removing material from the plant. Also, large
fans, pumps, and other equipment often operate 24 hr=day. There are a number of
approaches to control truck noise, including the use of proper exhaust mufflers and
enforcement of low-noise operating procedures. Also, the noise from the equipment in an
incinerator can be controlled by use of enclosures, or silencers, where appropriate
(particularly on the large fans).

XIV. INDUCED DRAFT FAN

The induced draft (ID) fan may be regarded as the heart of the system by effecting the
necessary gas flow without which the system could not operate. The ID fan draws the flue
gas from the furnace, through the heat recovery train and the air pollution control train, and
discharges it into the stack. The selection of the ID fan must be based on the maximum gas
volumetric flow, the gas temperature, and the estimated pressure drop through the
secondary furnace, boiler, air preheater, and the air pollution control train.
Heat and material balances can be used to estimate the design flow and temperature
data. Allowance must be made for further cooling through the air pollution control system
and the possibility of water vapor pickup if wet scrubbers are used. The estimate of the
pressure drop through the complete system requires an analysis of the flow resistance of
each component coupled with data based on experience in existing plants.
A.

Fan Types

Most induced draft fans are centrifugal fans although axial flow fans are also available.
Three basic types of centrifugal fans are commonly designated:
1.
2.
3.

Forward-curved blade fans with blades inclined in the direction of wheel

rotation
Radial blade fans with blades that are radial elements of the wheel
Backward-curved blade fans with blades inclined away from the direction of
wheel rotation

The most apparent difference between the three fan types lies in the pressure developed at
a given rotational speed. Where noise level is not a problem, the forward-curved fan
produces the greatest pressure (draft). Also, the volumetric capacity (the relative delivery
at a given static pressure of fans of the same wheel diameter) is highest with the forwardcurved type. However, experience in the United States suggests that fan noise is often a
significant problem.
The particular advantage offered by the radial blade fan is its ability to move fly-ashladen gases with less risk of blade deposits. Although the induced draft fan will handle
cleaned gas, the gas may still carry deposit-forming material. For this reason, the radial fan
is often considered the most suitable selection.

B.

Inlet and Outlet Connections

The design of the inlet connection between the furnace and air pollution control system
and the fan inlet can produce a marked effect on fan operation. Losses that arise from
improper inlet connections and duct work not only result in increased inherent pressure
losses but also create unbalanced, restricted, or whirling flow into the fan inlet. These
effects seriously reduce fan draft and efficiency.
Induced draft fans generally have shaped transition boxes as a feature of design. The
inlet duct is connected to the inlet box in a manner to ensure smooth, straight-line flow into
the fan inlet. Similarly, the fan outlet should be designed to gradually convert the high
outlet velocity pressure to static pressure as it discharges into the stack. Proper design can
provide further reduction in operating power and speed for a given fan.
C.

Fan Control

Volumetric gas flow and draft must be controlled to maintain balanced draft in the furnace
under the varying conditions of waste characteristics and burn rates. Automatic control is
very important for this function and is strongly recommended. Control methods generally
fall into three classifications:
1.
2.
3.

1.

Constant drive and fan speed with auxiliary device (damper) to modify volume
flow
Constant-speed drive with intermediate device between drive and fan to provide
adjustable fan speed
Variable-speed drive directly connected to fan, thereby providing capacity
control.

Damper Control

A sliding power-operated damper may be provided at some convenient point at the fan
inlet to throttle volume flow as needed. This method results in relatively heavy pressure
losses for a given amount of flow reduction and therefore higher power demand. This
method of control provides minimum first cost, maximum operating cost, wide-range
regulation, and simplicity of operation.
2.

Inlet-Vane Control

This method consists of a series of adjustable-position radial vanes or blades located at the
inlet of the fan. The vanes can be adjusted to various positions so that the entering gas is
given a change in direction or spin in the direction of fan rotation. This method offers a low
first cost and avoids the excess pressure dissipation of damper control, and the power
demand is correspondingly reduced compared with damper control. This is the most
common method employed on small incineration systems.
For induced draft fans, a louver-type vane control is commonly used at the inlet of
the inlet box attached to the fan similar to, but in place of, vane control, which is attached
directly to the fan inlet.
3.

Hydraulic and Magnetic Drive Control

The hydraulic method combines a constant-speed drive (motor or turbine) with an

adjustable-speed power-transmitting device to provide stepless adjustable control of fan
speed. The magnetic drive consists of a ring driven by a constant-speed motor, a magnetic
element rotating within the ring, and the ring connected to the fan. Both the hydraulic and

magnetic drive methods of control provide a wide-range, low-power consumption option,

having an effective response to control demand and with a first cost comparable with
variable-speed drives.
4.

Variable-Speed Drives

Fan-capacity control by variable-speed motors is not as common in power plant applications as the other applications discussed above, all of which use constant-speed motors.
However, variable-speed drives, although expensive, are quite effective in this service.

XV.

INCINERATOR STACKS

Several types of stacks or chimneys are used to discharge incinerator flue gases into the
ambient atmosphere. The primary engineering design issues include corrosion due to flue
gas, flow pressures, thermal stresses, forces due to wind and earthquakes, and weathering
conditions.
Stub stacks are usually fabricated of steel and extend a minimum distance upward
from the discharge of an induced draft fan. Stacks taller than about 5 diameters and less
than 30 m high are often referred to as short stacks. These are constructed either of
unlined or refractory-lined steel plate or entirely of refractory and structural brick. Tall
stacks are constructed of the same materials as the short stacks and are used to provide
greater draft than that resulting from the shorter stack and to obtain more effective
diffusion of the flue gas effluent.
Some metal stacks are made with a double wall with an air space between the metal
sheets. This double wall provides an insulating air pocket to prevent condensation on the
inside of the stack and thus avoid corrosion of the metal. The draft or negative pressure
within a modern incinerator is commonly provided with an induced draft fan. The need for
a fan (as opposed to natural draft) is occasioned by the high-pressure drop of modern air
pollution control devices, the depressed stack temperatures occurring when wet scrubbers
or dry scrubbers are used, and the control advantages realized with mechanical draft
systems.
Whenever stack gas temperatures are higher than the ambient, draft will be produced
by the buoyancy of the hot gases. The theoretical natural draft may be estimated by
!
1
1
PND 0:0342Pa Ls
"
2
Ta Ts
where
PND theoretical draft atm
Pa barometric ambient pressure atm
Ls height of stack above the breeching m
Ta ambient temperature K
Ts stack gas temperature K
The value used for Ts should be an average value taking into account heat losses and (for
tall stacks) the temperature drop on expansion (Eq. 3) as the atmospheric pressure declines
with increasing altitude
p p0 1 " 2:244 ( 10"5 H5:256

where p is the pressure at an attitude H (m) above sea level and p0 is the pressure at sea
level. As the stack gases pass up the stack, frictional losses reduce the fraction of the
theoretical draft available. Frictional losses may be estimated (418) from the following:
Fs

2:94 ( 10"5 Ls !u2 stack perimeter

Ts stack cross-sectional area

where
Fs friction loss atm
u! mean stack gas velocity m=sec
Stack perimeter and area are in m and m2, respectively.
In addition, some of the draft is lost to the velocity head in the rising gases. This
loss, referred to as the expansion loss (Fe ) may be estimated (418) by:
Fe

1:764 ( 10"4 !u2

Ts

where Fe is the expansion loss in atm.

Typical natural draft stacks operate with a gas velocity of about 10 m=sec. For forced
draft systems, 20 m=sec is more economical and provides a significant plume rise that adds
to the dispersion characteristics of the chimney (see Chapter 14).
Given that most incineration plants involve multiple boilers, each chimney shell
usually contains multiple flues (liners). Three or four liners can easily be installed in a
given shell. If future expansion is envisioned, considerable savings can be realized by
making the shell in the initial construction large enough to accommodate the liner serving
the new capacity. The multiflue concept allows for several chimney design options:
Circular and noncircular concrete chimneys
Steel stacks
Precast concrete chimneys
Special surface finishes and points
Relief line details
Brick or other insert panels
Provisions for stack sampling, flue gas monitoring
Access for inspection and maintenance
Aviation lighting
Lightning protection
Cast-in-place concrete chimneys: These comprise the majority of chimneys since they
provide superior resistance to environmental forces and are flexible as to overall height and
diameter. Generally, a cylindrical cross-section is preferred due to savings in forming and
structural strength, but architectural features are increasingly important. Departing from
cylindrical shapes demands a comprehensive structural analysis looking at static and,
particularly, the response to periodic dynamic loads such as from vortex shedding (wind
vibration).
Steel stacks: These stacks are limited in height and diameter, but their low cost and short
erection schedules make them attractive in some circumstances. Because of the low mass
of these stacks, they are more susceptible to wind vibration and interaction with vortices
shed from nearby stacks. The principal disadvantage of steel stacks is their size limitation.

This is important if multiple liners are envisioned since substantial clearances are desirable
in the shell-liner space (experience has shown that about 1 m is the smallest space required
to inspect and maintain the liner).
Precast chimneys: Precast chimneys have several attractive features (reduced installation
time, lower purchase price) that arise since most fabrication is done off-site where costs are
often lower. These units are cast using molds, and close tolerances can be maintained, thus
giving a good fit when the sections are assembled. Codes controlling the design and
installation of this kind of chimney are still being refined. However, this alternative can
possibly offer a lower cost option (especially for noncircular shapes).
Liners: Clearly, one of the most important issues relating to the selection of liners is
corrosion, especially if the gas temperature can go below the dew point of flue gas acids
(H2 SO4 ; HCl; HF. Most plants are planned for hot operation, and carbon steel and
independent brick liners are common. The liners should project above the shell a minimum
of one-third of the shell diameter or two-thirds of the liners diameter to minimize
downwash. A skirt should be provided where the liners project through the roof to protect
the air space from weather.
Brick liners provide high corrosion resistance but are vulnerable to earthquake
forces. In systems where the stack operates under pressure, one must control the escape of
gas from the liner with condensation and corrosion of the outer shell. One can pressurize
the liner-shell space with a separate fan, but this increases complexity and has a continuing
operating cost for power and maintenance. Usually, an acid-proof brick is used with an
acid-resistant potassium silicate mortar.
Steel liners are gas-tight and can readily be designed to cope with local earthquake
requirements. Generally, a minimum plate thickness of 0.6 cm is used, including a 0.16-cm
corrosion allowance. A stainless steel (often, 316L) is used for the top 1.5 diameters (or, at
least, below roof level) for corrosion protection, but carbon steel is used for the remainder
(418). If the stack is operated below the acid dew point, brick liners or chrome=nickel alloy
(solid or clad) is recommended.
Accessories: Numerous accessory features should be incorporated into the stack specification. These include access ladders and working platforms for use in inspections and
maintenance and in larger stacks, for stack sampling activity; utilities (especially electrical
services, perhaps a jib beam to lift equipment and supplies and lighting); two down-leads
(for nonmetal stacks) for lighting protection; and aviation obstruction lights according to
the requirements of the relevant aviation agency.
XVI. REFUSE-DERIVED FUEL SYSTEMS
Refuse-derived fuel (RDF) systems are incinerator furnaces where the waste is processed
prior to combustion with the objective of significantly reducing its level of heterogeneity.
The strategic concept of processing heterogeneous solid waste prior to combustion has
several potential virtues:
1.

2.

In the course of processing, portions of the waste can be recovered and recycled.
Thus, materials recovery can be an inherent partner in on-going or new resource
recovery programs.
The processing line can be designed to produce a more homogeneous fuel. This
should stabilize the combustion process, facilitate more precise combustion
control, improve burnout, produce a more stable steaming rate, etc. Since

3.

4.

separation processes can remove stones, glass, and wet material, the mean
moisture content of the waste can be reduced and the combustible content
enriched, thus increasing the heat recovery potential. Further, one would expect
that with better combustion control, excess air levels can be lower, thus reducing
the capital cost and many operating expenses for incinerator furnaces, boilers,
fans, and air pollution control devices.
Improved combustion should reduce air pollution emissions related to unburned
or incompletely burned combustibles. Further, because of the materials recovery
steps, emissions related to specific waste components (e.g., PVC) can be
reduced.
Because the processed materials are more regular in physical characteristics
(e.g., particle size), materials handling should be easier to automate and should
work better.

It was recognized that these benefits would come with some cost: the capital and operating
cost increases associated with the processing systems, some increase in the fraction of the
incoming waste that was sent to landfill, possibly some minor environmental problems
(especially noise and dust from shredder and conveyor operations), and, perhaps, some risk
of injury or equipment outage due to explosions. All of these anticipated costs and some
that were not anticipated have, indeed been experiencedand often at an unanticipated
degree of severity and consequence.
In the development and implementation of technology for mass burn facilities, the
vendor firms participating in the market generally offer complete chute-to-stack systems
based on conservative, repeat application of well-proven technological features. Most
aspects of the mass burn technology were extensions of combustor concepts developed and
proven in burning coal, lignite, and waste fuels such as bagasse. In contrast, many RDF
plants incorporate innovative process flowsheets and experiment with untested and
unproven items of equipment in the critical RDF processing and handling areas. The
extensive body of design data that supported the early mass burn incinerator designs was
lacking for RDF systems. Further, the lack of overall system vendors for RDF-based
systems reduced the feedback of field experience into process development.
The often-demonstrated characteristic of refuse to profoundly stress, clog, abrade,
corrode, and otherwise challenge process and handling equipment made extrapolations
from prior experience with other materials uncertain and, quite often, disappointing.
Further, the prime RDF processing devices were connected together with materials
handling systems where critical design parameters were not well understood and where
refuse behavior often was found to be quite different from that of reference materials
used as the basis for design.
The consequences of these facts and influences was repeated in many if not all
plants: long and politically painful startup programs requiring considerable rework of
conveyors and processing equipment. This was accompanied by increases in facility
capitalization, higher-than-anticipated operations and maintenance costs, and, often,
derating of the combustor due to limitations in processing and=or feeding systems.
Also, from time to time, many plants experienced explosions, some with lethal consequences, from ignition of combustible vapors released in hammermilling of volatile
hydrocarbon (e.g., gasoline) containers.
Despite these challenges and failures, a maturing technology ultimately emerged. By
the early 1990s, package RDF processing facilities with performance guarantees could

be procured to generate specification RDF. A vendor community offering total RDF-based

systems was developing. RDF-only and co-combustion of RDF with other wastes and=or
coal were being practiced in several incinerator boiler and utility power plants throughout
the United States. The basic technology had become generally available. The primary
concerns limiting implementation of RDF-based projects had shifted from one of high
perceived technical risk and problematic operability to the more conventional project
development decision criteria: the ordinary issues of cost, permitting, siting, and the like.
A.

RDF Processing

The underlying objectives of incineration using RDF (materials recovery and production
of a homogeneous fuel) are achieved in the processing line. It is here that materials for
recycling are recovered (on picking belts by human workers or by automated processes
that exploit the physical or chemical characteristics of target refuse components). Before or
after the removal of materials, size reduction and size separation take place. In all cases,
the RDF facility is heavily involved in materials handling equipment to include a spectrum
of loaders, conveyors, hoppers, and feeders. The perverse nature of refuse and its
components with conventional materials handling processes has been the major problem
with RDF plants.
Table 13 lists several of the common technologies used in the separation of refuse
components in the course of RDF preparation. In addition, size-reduction and sizeseparation equipment are used. These elements of RDF incineration systems are described
in Chapter 7. Typical final characteristics of the RDF (532) are shown in Table 14.
B.

RDF Combustion Systems

The combustion concepts used in burning RDF fall into two categories: RDF-only
combustors and combustors where RDF is burned as a second fuel along with coal,
wood waste, or other materials. Since almost all processing concepts include size reduction
and the removal of massive metal material, stones, much of the glass, and the very wet
Table 13 Materials Recovery Methods
Method

Component affected

Visual

Aluminum cans
Corrugated paper
Newsprint
Glass containers (by color)
Plastic containers (by type)
Auminum
Ferrous metal
Lights (paper, cardboard)
Heavies (glass, metal, stone, very wet)
Glass (green, brown, flint)
Plastic (unchlorinated from PVC)
Separate by particle size to reject glass and sand and enrich
fuel value. Also, separate over-size for recycle to size
reduction

Eddy current magnet

Magnetic separation (belt and drum)
Air classification
Photodection of color
X-ray emission by chlorine atoms
Trommels and disk screens

Table 14 RDF Quality Impacts of Preprocessing Strategy

Process sequence
Without front-end recycling
Mass burn (no preprocessing)
RDF via shred, Fe recovery
RDF with classification, etc.
With front-end recycling
Mass burn (no preprocessing)
RDF via shred, Fe recovery
RDF with classification, etc.

RDF yield %

Landfill %

Ash %

Kcal=kg

100
93
83
to 70

0
7
17
30

23.6
19.9
11.7
to 8.9

2675
2860
3135
3240

100
93
85
to 71

0
7
15
29

19.6
17.2
9.9
to 6.6

3060
3275
3515
3605

Source: From (532).

wastes, the remaining fuel is well suited for partial or complete combustion in which the
particles of RDF are suspended in gas flows. RDF is poorly suited for grate burning, as it
tends to blind the grates and inhibit air flow. Consequently, the types of combustion system
used for RDF have focused on the spreader stoker (semisuspension burning) and several
embodiments of full suspension firing.
1.

Spreader Stoker Firing

The spreader stoker furnace uses a single, flat traveling grate. The grate moves at a slow,
constant rate much as a conveyor belt. Air is supplied through the grate from one or more
undergrate plenums. The grate elements are designed to develop a relatively high-pressure
drop at operating underfire air rates, which results in a uniform air distribution over the
entire grate surface. Waste supplied to the furnace is typically shredded such that about
95% passes a 10.2-cm top size. The grate elements are generally fabricated of ductile iron
for resistance to mechanical shock.
Wastes are charged to the furnace using several air-swept feeders mounted on the
front face of the furnace wall. The feeders cast the feed over the fire to land on the grate
near the rear wall. The grate moves the waste back toward the front face and discharges ash
just below the feeders. All but the very wettest RDF feedstock is dried and ignited in its
flight across the furnace through the flow of hot combustion gases rising from the grate.
After landing on the grate, the waste continues to burn. For design purposes, it is often
assumed that 40% to 60% of the heat release takes place in suspension and the remainder
on the grate. The bed depth is maintained between 15 and 30 cm to foster good burnout
and to maintain good grate pressure drop (and, as a consequence, good air distribution).
Typical grate heat release rate and furnace volumetric heat release rates in several U.S.
RDF-based refuse burning applications are summarized in Table 15 based on the total
RDF fuel energy input (515). The grate heat release levels are somewhat lower than might
be found in, say, a bark-burning furnace.
It is particularly important with spreader stoker burning to get good distribution of
fuel. The RDF feeder chutes are swept with a pulsing flow of air (variation in both quantity
and pressure at several cycles per minute). The air flow sweeps the bottom of the spout and
floats fine, low-density material far into the furnace. Under each spout, a bank of high-

Table 15 RDF Grate and Furnace Heat Release Rates in U.S. Installations

Facility
Palm Beach (FL)
MERC (ME)
Elk River (MN)
SEMASS (MA)
PERC (ME)
Colombus (OH)
SPSA (VA)
Honolulu (HI)

No.
units

TPD per
unit

Furnace
volume
m3

Grate
area
m2

Typical
CO
(ppmdv)

Kcal=hr
per m3

Kcal=hr
per m2

2
2
1
3
2
4
4
2

818
273
545
818
319
455
455
809

929
201
478
1193
318
260
500
751

58.5
20.4
37.1
55.7
27.3
31.8
37.1
36.8

55
90
125
140
200
300
400
100

124,843
188,208
132,425
84,507
150,978
228,357
129,577
116,906

1,980,625
1,949,496
1,709,333
1,808,397
1,758,308
1,869,582
1,748,180
2,387,707

Source: From (575).

pressure air jets act to fan the stream of air-suspended RDF to improve the side-to-side
distribution. The pneumatic feeders are described and shown in Section VI above.
Among other issues, poor fuel distribution leads to elevated CO emissions, even to
the point where some U.S. RDF boilers burn natural gas or oil through burners mounted
just above the firing rate and operating at about 20% excess air to burn out CO. Data on
typical CO levels are shown in Table 15. The burner firing rate is adjusted manually or
under the control of the CO continuous-emission monitoring system. High CO spikes are
also observed when feeder plugging occurs.
Although the RDF preparation step can produce a fuel with considerably greater
homogeneity than raw refuse, the benefits of homogeneity require a steady feed rate.
Achieving a uniform feed rate has not proven easy. Hangups and blockage occur in almost
all systems. This results in irregular feed rates, irregular heat release, and irregular
steaming rates. The best performance in feeding has used a run-around concept
where the mass flow rate of refuse to the boiler exceeds the firing rate. Thus, a substantial
quantity of the RDF is recycled.
Overfire air jets are, perhaps, more important for spreader stoker firing than with
mass burn units. New units are designed for 50% of the total air supply over the fire with
normal operation at 40%. The nature of the firing process gasifies a relatively large fraction
of the RDF volatile matter at a high elevation in the furnace. Also, a substantial mass of
solid matter is suspended in the flue gases. The high-pressure overfire air jets discharging
above the arc of incoming RDF are critical to induce the intense mixing needed to achieve
burnout of suspended particles and volatile matter. It has been speculated that the reason
that the polychlorinated dioxin=furan emissions of RDF furnaces are observed to exceed
those of mass burn systems is due to incompletely burned organic precursors that are
part of the solids carried out of the primary combustion zone.
Because of the substantial gasification of the refuse while in suspension and the
relative homogeneity of the material, the flue gas environment near the side-wall does not
show the alternating oxidizing and reducing condition that produces side-wall corrosion in
mass burning furnaces. Data at the City of Hamilton, Ontario, RDF furnace (415) showed
no significant side-wall attack after four years of operation at overall excess air levels as
low as 40%. As a consequence, no refractory cladding appears to be needed. This increases

the heat absorption efficiency of the furnace waterwalls and reduces the boiler investment
cost.
2.

Suspension Burning

The burning of RDF is suspension has been effected in two ways:

1.
2.

In solid fuel boilers using feeding methods that are similar to those used to fire
pulverized coal in suspension and in vortex furnaces
In fluid bed combustors

In each case, secondary fuels (coal or other waste streams such as wood scraps, sawdust, or
agricultural wastes or industrial liquid waste) are often burned at relative heat release rates
that range from zero to many times that of the RDF.
a. RDF and RDF-Coal Burning in Suspension-Fired Boilers. Co-burning of refuse and
coal is significantly different from burning either fuel alone. With co-burning, the
combustion environment must be tailored, to a degree, to the limitations imposed by the
poorest fuel. The furnace temperatures will approach that of the dominant (heat release
basis) fuel. In retrofit situations, the physical combustor characteristics (height, residence
time, heat release patterns, boiler-tube locations, etc.) are inflexible and some compromises
will be forced on the operator. Clearly, if either fuel presents failure modes (e.g., low ash
fusion temperature or slow combustion), the situation may be such that problems will,
necessarily, arise. However, the success to date in refusecoal co-burning suggests that the
problems can be solved, mitigated, or lived with. There will, however, be a greater level of
maintenance, increased outage frequency, and many aggravations related to plugging,
fouling, and the like.
As the particle size of a solid fuel diminishes, behavior in a combustor approaches
that of gaseous or liquid fuels. The particles volatilize rapidly. The small char=ash particle
that remains has a large surface area, thus facilitating rapid oxidation. To approach this
idealized combustion state, coal is pulverized to approximately 200 mesh: only slightly
grittier than face powder. In this form, powdered coal may be entrained in an air flow,
swept up in an ignitable cloud, and burned to completion in only a few seconds.
In comparison to lump coal that fractures and shatters quite readily, refuse does not
pulverize easily. The most important fuel components of refuse are paper and plastic. Both
deform and stretch under impact and degrade slowly (and with high energy absorption) in
crushing, cutting, or tearing environments. This means that the ultimate particles will be
large. The economic practical lower limit on particle-size reduction is approximately 2 cm.
Processing costs increase rapidly as the top size decreases below this level.
The top size of the RDF used in suspension burning has varied from as small as
0.95 cm to as much as 6.35 cm. The cost of shredding increases rapidly as the mean
particle size decreases. However, larger particles have longer burning time. One should
avoid carryover of still-burning RDF because (1) the material can generate clinkers during
burnout in the ash hoppers under the boiler passes or, (2) if applicable, still-burning
sparklers cause pinholes in fabric filters. Therefore, the goal has been to find the largest
top size that gives acceptable combustion and minimum carryover of burning material
from the primary furnace.
A maximum particle size between 2.5 and 3.8 cm appears optimal. Tests by the
Potomac Electric Company with RDF shredded to 95% minus 1 cm appear to have reached

the point where the burnout grate is not necessary. The high cost of processing to this
degree makes this strategy unattractive.
In addition to combustion problems associated with mean particle size, a second
problem area arises because paper and other refuse components can absorb and=or carry
considerable moisture. This contrasts with coal, which can be largely dried in the air-swept
pulverizer. The RDF portion that is wet with surface moisture can tolerate several seconds
of intense radiant heating before thermal volatilization begins. If the cross-section is thick,
the time required for conduction heat transfer and pyrolytic breakdown of the organic
matter further delays burnout. This means that the general time frame to achieve full
combustion for a significant portion of the refuse fuel will significantly exceed that of coal.
The result is a higher carbon loss (unburned carbon heating value in the ash residues) and
unfavorable decreases in fly-ash resistivity (important in plants using electrostatic
precipitators).
The energy density (kcal=m3 ) and specific volumetric air rate (m3 of air per m3 of
RDF) for RDF and coal are significantly different. Also, the fuel handling tendencies (ease
in maintaining a uniform feed rate, potential to hang up or build up on projections, at
transfer points, etc.) of coal and RDF are very different. Because of these significant
differences and because the coal is usually blown into the boiler directly from the coal
grinders (without any buffering hopper where coal and RDF could be blended), it has been
found that separate burners for RDF and coal are to be greatly preferred over a mixed-fuel
burner concept. Consideration has been given, however, to the use of a burner fabricated of
concentric pipes to allow firing of both fuels from the same burner port. In general, the air
rate for the refuse burner is kept constant irrespective of the solid feed rate; otherwise, the
RDF will begin to settle out, build up, and block the burner piping.
Achieving complete burnout of co-fired RDF with acceptable processing costs
requires the installation of a burnout grate at the bottom of the primary furnace. The grate
not only helps to achieve substantial burnout (realization of most of the heating value of
the waste material), but it also eliminates the problem of fires from the accumulation of
combustible waste material in the ash handling system. The burnout grates used
successfully to date are simple, flat dump grates although several operators have modified
the size and shape of the air holes (increasing the hole area and changing them from
circular to slotted).
The grates are supplied with a small amount of air for the combustion of material on
the grate. This undergrate air flow has been shown to be necessary to achieve reasonable
burnout. The furnace waterwall near the dump grate is exposed to alternating eddies of
oxidizing and reducing conditions. These changing gas conditions lead to rapid tube
wastage. A simple and effective solution has been to supply a small air flow to the sidewall near the bottom of the boiler (an air curtain along the wall) to ensure oxidizing
conditions at the wall.
Corrosion is a major concern when contemplating co-burning of refuse with coal in
high-temperature, high-pressure boilers. The corrosion process of greatest importance is
high-temperature, chlorine corrosion. This corrosive process involves attack of the
superheater and boiler-tube surfaces in refuse incinerators where the fireside metal
temperature exceeds about 425! C. In the co-burning case, one has several factors that
appear to mitigate the corrosion problem: significant dilution of chlorine; significant
neutralization of hydrochloric acid by alkali in the coal ash; and significant counteraction
of the chlorine corrosion reactions by the high concentration of sulfur compounds. The net

result is that there have been no reported cases of severe tube corrosion in refusecoal
co-burning situations.
Opinions vary among operators whether it is preferable to put the RDF in through
burners mounted at the top of the burner array or closest to the burnout grate. High entry
applies the most intense radiative and convective heat flux to the incoming RDF, thus
hastening its final drying and volatilization, but gives the shortest residence time prior to
the particles entering the tube banks. The latter strategy avoids blowout of unburned RDF
but, probably, increases the dependency on grate-burning processes.
Refuse constituents (especially bottle glass and other mineral residues such as the
high-phosphorous ashes from grasses, food, and brush) generally have a low ash fusion
temperature relative to coal. Here, low is taken in the context of both the initial
deformation temperature of the refuse ash material as it compares to coal ash and the
expected high temperature of the coal flame. This can cause problems with slag
accumulation, can cause slag falls (with damage to the floor of the boiler passes), and
can severely stress the equipment for bottom ash handling.
Most U.S. utilities have operated suspension-fired boilers on mixtures of RDF and
coal rather than RDF only. The RDF has been limited to a maximum of about 15% of the total
furnace heat release rate. In view of the relative heating value and ash content of
coal and RDF, this corresponds to about 30% of the total mass input and ash output.
This also reflects the limits permitted by the U.S. EPA concerning the percent of RDF that
can be burned before the air pollution emission requirements shift from those of the coalburning utility industry to that of municipal incineration. Since the latter incur the need for acid
gas control, monitoring for dioxin compounds and other pollutants, the 30% mass rate level
remains as the maximum average firing rate where there is significant experience.
The vortex (cyclone-type) combustion technique is an alternative to the suspension burning methods described above. In the vortex furnace, the airfuel mixture is
injected at high velocity at an angle tangential to the inner wall of a large water-cooled
burner mounted in one wall of the boiler furnace. The excess air is restricted to achieve
high flame temperatures. The ash residues are thrown to the wall, where they accumulate
and melt under the intense radiative and convective flux and are tapped from the furnace as
a liquid slag. This furnace configuration has been in disfavor in recent years due to its high
NOx emission profile so no new boilers with this burner design are planned. Data are
limited, but the results appear promising as an alternative in existing vortex furnace
installations.
Regardless of these apparent problems, refusecoal co-burning in the suspension
burning mode has been successfully accomplished in the United States by several utilities
and=or large industrial boilers on a long-term (up to 17 years) basis. Most of the boilers
firing RDF have practiced tangential firing (wherein the fuel is suspended in an air
stream and fired from the corners of a square cross-section furnace through burners aimed
tangent to an imaginary cylinder concentric with the furnace centerline). However, several
boilers have been front-fired: where the air-suspended RDF and coal are fired from an array
of burners mounted on one wall of the boiler. The boiler plants co-firing refuse and coal
include
Union Electric (St. Louis, MO)
Ames Municipal Light Co. (Ames, IO)
Lakeland Municipal Light Co. (Lakeland, FL)
Madison Gas and Electric (Madison, WI)

Tangential
Tangential & front
Front
Tangential

Rochester Gas and Electric (Rochester, NY)

Potomac Electric Plant
Northern States Power (several)
Eastman Kodak Co. (Rochester, NY)

Tangential
Tangential
Tangential, vortex & front
Tangential

Several of these owner=operators continue to make investments and, importantly, reinvestments in co-burning facilities. Co-burning has not been easy, but the environmental,
financial, and=or political benefits appear evident to these owners.
b. Fluid Bed Burning. A fluid bed incinerator=combustor for RDF involves a cylindrical
or rectangular chamber containing coarse sand or similar bed material through which a gas
is passed to cause the sand to bubble and boil much as a liquid. The fluid bed concept was
originally developed as a solids-to-gas contacting device for catalytic operations in the
petroleum field. The applicability of the principles of fluidization were soon extended to
drying, ore processing, and, ultimately, waste incineration. The development of technology
to add boiler surface area removing heat from the bed walls and bayonet tubes removing
heat from within the bed itself provided the key to the use of the fluid bed as a generalpurpose combustor for solid fuels such as coal, refuse, and wood.
The application of fluid bed technology to refuse burning has been accomplished in
several plants in Europe and Japan using circulating fluid bed technology. (See Chapter
11 for more details on the combustion concept.) Here, one accepts the slow burning time
of the solids and compensates by (1) increasing the flow velocities to transport all of the
solids out of the bed and (2) putting a cyclone or other device in the leaving gas flow to
capture and recycle (or discharge) the solids. In time, an acceptable level of burnout is
achieved.
A second approach to the use of fluid bed technology involves modification of an
existing coal furnace (suspension-fired or stoker-fired) to add the distribution plate, highpressure air supply, sand management, and other features of a bubbling fluid bed. (See
Chapter 9 for more details on the basic combustion concept.) RDF, coal, wood, or almost
any other feedstock that is compatible with a reasonable overall energy balance can be fed
to the bed. A critical requirement is to incorporate features that can adequately handle the
segregation and discharge of the noncombustible, tramp material fed to the system.
Also, it is necessary to effect heat removal from the bed. Heat removal allows control of
bed temperatures and avoids ash fusion and bed defluidization. Facilities meeting these
requirements are operational in La Crosse, WI, and Tacoma, WA to burn municipal RDF.
In addition several other plants use the same process concepts in industrial waste
applications.
The La Crosse facility (417) was a reconstruction of a spreaderstoker-fired coal
boiler. The fluid bed components were, in essence, a replacement for the stoker and are
used to burn RDF and=or hogged wood waste. The Tacoma fluid bed (416) is a separate
combustor, rigged external to the boiler and discharging hot combustion gases into two,
25-MW existing boilers (previously, a front-fired pulverized coal system). For Tacoma, the
existing boiler became, in effect, a waste heat boiler designed for high-ash fuels. Tacoma
burns coal and hogged wood waste as alternatives or together with RDF. Both plants have
operated more than 5 years with acceptable on-line availability and economic viability.
RDF for the Tacoma fluid bed system is prepared by shredding to a 1.9 by 10 cm
product using a horizontal shaft hammermill. La Crosse begins with a flail mill and disk
screening followed by a secondary hammermill shredding. Ferrous metal is removed from
both products, and heavies are separated with air classification techniques. The RDF

materials contain between 2% and 10% ash and have about 3000 kcal=kg higher heating
value. The preparation plant achieves about a 50% yield of RDF vis-a-vis the received
refuse. Some compaction occurs in hauling and handling the RDF, but the material
refluffs well on handling.
In Tacoma, RDF is dumped in a special area in the feed house: a large, covered,
metal building-type structure with four bays for the holding and feeding of the RDF and
wood waste. The RDF is moved about, as required, using a large front-end loader. Wood
waste is received in large, open-top, semi-trailer truck bodies and, with an automatic, tipup-type unloader, is emptied into a holding hopper. The wood is then conveyed by belt
conveyor to one of several bays in the feed house.
Each bay is provided with a drag conveyor cantilevered over the waste piles. The
bite of the conveyor into the pile is adjusted by an operator observing the pile and
controlling the fuel reclaim operation from a second-story office overlooking the waste
piles. The wood and=or RDF is dragged to the back wall of the feed house by the conveyor
flights and drops onto a belt conveyor running the length of the building. The conveyor
becomes inclined as the end of the building is approached and the feed is elevated to a
holding and feeding bin on the roof of the boiler house.
At Tacoma, the wood and RDF are fed from the holding hopper atop the boiler
across a vibrating feed table and through a metering feed system (a rotating vane that
provides both a metering action and an air lock between the combustor and the feed
storage). The material then discharges from four points above the bed. The RDF is
regarded as a good fuel of reliable characteristics. The most serious problem in its use is
irregular feeding (tends to hang up and then spurt, hang up and spurt, etc.). The
compressibility of the RDF leads to packing with the slightest reduction in cross-section
of a flow path. The average fuel firing corresponds to:

Fuel

Percent heat release

Mass rate (kg=hr)

Heat rate
(1000 kcal=hr)

Coal
Wood
RDF

50%
35%
15%

13,600
34,000
12,700

92.5
64.8
27.7

Each Tacoma fluid bed combustor is equipped with two cyclones to capture and recirculate
incompletely burned solids. Each fluid bed includes 471 m2 of half-submerged bed tubes
to remove 58 million kcal=hr of excess heat while evaporating 137,000 kg=hr of water
from the fluid bed. In comparison with the concept used in La Crosse (where the bottom of
the boiler was cut off and replaced with the fluid bed distribution plate), this results in a
higher capital cost and a reduction in the maximum steam production below the original
design level. Design steam production is 240,000 kg=hr at 29 bar, 400! C.
The Tacoma bed is about 0.9 to 1.07 m deep (0.61 m slumped depth). Most of the
5555 kcal=kg sub-bituminous coal (10% moisture, 30% volatile matter) burns in the bed at
790! to 815! C. The wood burns on the top of the bed. Refuse and coal fines burn in the
freeboard at 870! to 900! C.
Corrosion of the old (>50 years) Tacoma coal boiler has not been a problem,
although attack of weldments in the 309310 stainless steel bed tubes has been observed.
Flame spraying of protective materials, the use of extra thick (Schedule 320) tubes for the

submerged portion of the bed tubes, and the addition of fins to the bed tubes to break the
bubbles appear to have resolved this problem. Bed erosion problems at La Crosse were
resolved by the addition of split pipe shields to the bottom half of the bed tubes.
Some buildup of clinker in the ductwork and in the superheater has occurred despite
the cooling of the furnace gases with the bed tubes. It has been suggested that the high
alkalinity of the wood ash may act as a low-melting binder for ash. Clinker buildup has
been a problem since the massive ash accumulations break off, fall into the bed, and bend
or crush the bed tubes.
The fluid bed concept has been successful in both extending the operating life of the
existing boilers and providing economic solid waste disposal. Addition of limestone to the
fluid bed has been effective in control of HCl and sulfur dioxide. Operating availability has
been acceptable.
XVII.

INSTRUMENTATION AND CONTROL

Effective instrumentation and controls are key to maintaining plant operation at all times
within its design envelope, with desired ash burnout and within environmental regulations.
Even if an abundance of labor is available for manual operation and control, it is impossible
to provide timely and effective coordination and supervision of the many operating
variables without a minimum degree of instrumentation and automatic controls. The
degree of automated instrumentation is a decision up to the designer, who must balance the
level of operator skill and manpower resources versus the benefits and disadvantages of
automation. Incinerators demand steady operation, and quick responses and actions are
often required to effectively respond to unexpected process upsets as the waste quantity and
characteristics vary. Adjustments must be made to the process variables to meet the plant
objectives. The plant operators can readily make some of these adjustments while others
are more suited for automation. For some applications within the plant, automated controls
will actually do a better job than an operator could do manually.
Monitoring instrumentation that samples the current value of critical process
parameters provides plant operators or automated control systems with the information
needed to make the necessary adjustments. The plant operating concept can provide for
these adjustments to be made manually, but this assumes the continual presence and
attention of well-trained, dedicated, and knowledgeable operators. Some form of automatic
measurement and control may be provided to take critical process operations out of the
hands of the operator and respond directly and instantaneously to plant upsets and
operating variations. The sections that follow indicate what is regarded as the appropriate
control system, consistent with industry practice for modern MWC systems.
A.

Instrumentation and Control System Design Approach

The design effort for plant instrumentation involves several steps leading to a functional
system that can effectively and efficiently run the plant. The first step is to prepare a
conceptual functional instrumentation specification. This specification should be prepared
based on the overall scope of the project once the process flow diagram for the plant has
been established and includes an identification of all key systems and equipment to be
included in the overall scope of the project. All components within the system as well as
the basic control parameters should be identified. Once agreed upon, the conceptual
specification should be developed into a design specification that outlines how the scope is

defined in the conceptual instrumentation and control specification and identifies those
responsible for each component. This specification includes hardware layouts, process
flow diagrams, detailed logic diagrams, and data recording and reporting requirements.
More specifically, the design specification
Identifies systems designed in-house, those systems that will be subcontracted out,
and those major systems that will be supplied as a package that includes complete
instrumentation (e.g., refuse crane controls, air pollution control equipment).
Identifies interface requirements and defines scopes.
Coordinates the use of numbering, symbols, and instrument identification.
Completes preliminary process flow diagrams, P&IDs, and control descriptions.
Consideration must be given as to how the system will be controlled and, also,
fail-safe control logic for safety and protection of equipment. The designer must
ask what will happen if each piece of equipment fails and develop appropriate
interlock and fail-safe logic to protect personnel and equipment.
Prepares instrument lists and logic diagrams for system interlock, sequencing,
control, and alarm systems.
Prepares data sheets of instrumentation components and functional specifications for
procurement.
Sizes all instruments, control valves, relief valves, and metering devices.
Prepares supporting documents for instrument installation.
Reviews instrument equipment vendors loop and wiring drawings and schematic
drawings by packaged unit vendors.
Prepares instrumentation location and hookup drawings.
B.

Process Measurements and Field Instruments

Process measurements obtain necessary data to monitor and control plant operation. These
include temperature, pressure, flow, position, level, and flue gas concentrations. The choice
of a local readout device or an electronic instrument with transmitted signal and its
location should be made with care paid to process considerations and the importance of the
parameter, requirements for recording, and the overall control philosophy.
Instruments used to monitor process variables range from relatively simple devices
such as thermocouples for temperature measurement and pressure gauges, to position
indicators and limit switches, to more complex instruments for flow measurement and flue
gas pollutant concentrations. It is important to select the appropriate measurement device
for the particular application, purpose, and operating environment. Compatibility with the
temperature, humidity, dustiness, and other characteristics of the operating environment is
especially important in incinerator applications. Due to the aggressive, corrosive, and
abrasive nature of waste, waste dust, and the combustion products of waste, particularly
rugged equipment is needed or accuracy will degrade and instrument life will be
unacceptably short. Close consultation with manufacturers of the devices is necessary
during the design process.
It is equally important to determine the appropriate level of instrumentation and
control necessary or desired for a particular application. Process measurements can be
displayed at the instrument itself or using a transmitter sending a signal to a local control
panel and=or the central control room. Display-only devices are referred to as indication
versus transmitting devices that send the measurement signal for use in open- or closedloop control. An open-loop controller is operated manually from a control station, while a

closed-loop controller controls a variable measurement automatically from a preset set

point. The incineration plant will have an abundance of instruments used for indication,
open- and closed-loop control, or combinations thereof.
The basic parameters used to operate an MWC include combustion temperature,
under- and overfire air damper settings, combustion air flow and temperature, grate speed,
furnace draft, steam system parameters (water level, steam temperature, and pressure), and
flue gas concentrations of oxygen and regulated pollutants (e.g., CO or SO2 ). Indication of
these measurements should be provided locally in most cases. However, for critical values
and data that are used for subsequent control purposes, data need to be transmitted from
the field device to either a local control panel and=or central control room. Then,
information can be either used by the operator for control (open-loop control) or
transmitted for subsequent automatic control (closed-loop control). Table 16 is a listing
of the main operating data and the associated instrumentation typically employed in an
incineration plant.
Some process measurements are used both to monitor and to control using either an
open or closed control loop. Typically, motorized or pneumatic dampers and valves and
motors coupled to such items as pumps, drives, conveyors, and fans are controlled. The
combination of the instrument, transmitter, control logic, and device achieves the desired
control function. For example, a low reading for the water level in the steam drum may be
indicated by a level gauge coupled to a float switch that in turn sends a signal to a valve
that increases the feedwater rate until the desired level set point is achieved. Many types of
instruments, transmitters, controllers, and recorders are available to the designer for
application to an incineration system.

C.

Control System Levels

Three levels are involved in the control system. Information and data may also be
transmitted or shared between the various control stations as well as control functions
themselves facilitated from one or a combination of control stations.
1. Level 1: hardwired controls including local controls, controls at motor control
centers, and all equipment and protective interlocks. The inter-locks are part of the
hardwired controls so that they are active in all modes of control.
Table 16 Principal Operating Data
Parameter
Mass of waste
Combustion and flue gas temperature
Damper settings
Combustion air=flue gas flow
Grate speed
Draft
Flue gas analysis (O2 , CO, SO2 )
Steam system properties
Tank levels
pH

Instrument
Load cell
Thermocouple
Position indicator
Differential pressure gauge=flowmeter
Tachometer
Draft gauge
Flue gas analyzer and CEM analyzers
Thermocouple, pressure gauge, flowmeter, sight level
Level gauge, float
pH meter

2. Level 2: local panel level. The packaged equipment control systems include
these in some form. Other systems may include local control panels unless the controls for
those systems are implemented in the plant control system.
3. Level 3: control room. Equipment here can either be computer-based workstations or a master panel with recorders, controllers, switches, and indicating lights.
1.

Local Control Stations

Local control stations are

Always located near the equipment
Normally used for maintenance rather than for process control
Sometimes used for simple control by an operator in the field when visual contact
with the process or equipment is essential or the necessary adjustments are
infrequent or noncritical to the overall operation
Local control stations contain no logic or programming; however, data may be
transmitted to local control panels or the central control room as required either
for monitoring or use in a control loop. All local control stations should have
on=off indicating lights and other indicators of system performance to aid the
operator in the field.
Examples include on=off switches for motors so that they can be shut down for
maintenance, manual valves and dampers and independent batch operations such as the
filling of tanks, and manual preparation of chemicals.
2.

Local Control Panels

Local control panels contain programmable controllers, switches, relays, or other control
logic receiving signals from field instruments and performing control functions. Two types
of local control panels are used:
1.
2.

Control panels provided for major systems.

Control panels provided with smaller packaged equipment. These panels also
generally interface with the central control room or may in some cases be
physically located in the central control room.

Major system suppliers (e.g., the air pollution control supplier) provide sophisticated
equipment comprising a complete control system. The system includes the complex
control logic necessary for automated operation, incorporates necessary safety interlocks,
and interfaces with other systems. These panels are generally based on programmable logic
controllers (PLC) and can carry out complex control functions. The panels are generally
located in the central control room along with panels for other major systems.
Smaller equipment can be provided with PLCs switches or relays. These panels are
usually located near the equipment. Examples include a lime slurry preparation system, the
refuse crane management system, packaged wastewater treatment plants, continuousemission monitoring systems (CEMS), ash handling systems, etc. Depending on the
application these panels may be controlled.
Locally in the field
From larger control panels
From the central control room

Local control panels may also provide indication or data information to other control
panels.
3.

Central Control Room

Major data points and controls are monitored, controlled, and recorded in the central
control room. Measurements throughout the plant are fed back into the control room and
input to a central computer=data logger. Critical measurements such as feed rate, primary
and secondary chamber temperature, and O2, SO2 , and CO concentration are usually
continuously displayed while others may be recalled by the operator through the computer.
Significant open-loop controls pertaining to the overall process should be controllable
from the central control room by the operators. Closed-loop controllers are monitored in
the control room and may be manually overridden or the set points adjusted at the
discretion of the control room operator. In addition, the computer performs data logging,
alarming, as well as storing of historical information.
4.

Data Logging

Critical measurements should be recorded to provide an archived record of the status of all
plant equipment for subsequent analysis of plant failures or upsets and to verify attainment
of performance guarantees under the provisions of the Service Agreement. Further, the
data may be submitted to (or periodically accessed by) regulatory authorities to establish
that permit requirements are being met. Also, the data base collected in the data logger
provides a basis for future design improvements.
Records derived from plant instrumentation may be hard copy made by the
recording instruments themselves on circular or strip charts, or computer data storage. The
latter can provide real-time data trends to guide plant operators and the periodic recall of
historical data. Also, the computer data base can be accessed to readily generate reports of
overall plant performance. Computers offer the most accurate and efficient means of data
collection and data processing. It is strongly recommended, therefore, that a computer data
logging system be incorporated into the plant design.
D.

General Control Philosophy

The general control concept for an MWC is to operate the plant from a central control
room. This optimizes operating efficiency and overall plant safety and minimizes
environmental impact. Process controls that require infrequent adjustment or are small,
batch-type operations can be operated in the field and manually adjusted as needed. This
section provides typical control descriptions and parameters necessary for proper operation
of the plant as well as an indication of the type of control recommended.
1.

Automated Refuse Weighing

The measurement and recording of the refuse weight are often critical parameters in many
Service Agreements. Also, the mass input of wastes is often the measure by which an
incinerator operations firm receives payment. From a process analysis standpoint, waste
throughput is one of the most basic parameters needed to monitor system performance in
terms of combustion capacity.
Refuse is usually weighed through the use of a truck platform scale. The weighing
and recording system can be obtained readily from scale vendors. The software can be
customized to meet the particular needs of the project even to the degree that the

information collected is directly tied into the accounting and billing systems. When
specifying the scale and recording system, the particular needs of the project must be
known and the scale must be specified to handle the largest vehicle type that would access
the facility. The computerized weighing system, if needed on a project, should include the
recording of the following, at a minimum:
Vehicle identification
Vehicle tare weight
Vehicle laden weight
Vehicle load (by difference)
Time
Date
Daily, weekly, monthly, quarterly, and annual summaries
If billing is not needed on a tons-processed basis, a facility scale and scale house may not
be necessary, but a system for measuring the feed rate to the incinerator is highly desirable.
This typically involves load cells on the refuse cranes that weigh and record each grapple
load as it is being charged. This function should be incorporated into the crane
specification. The refuse mass measurement system should record and transmit to the
central control room at least the following:
Crane identification
Date
Time
Charge weight
Hourly, daily, and weekly summaries
Incorporating a scale and crane recording system is most desirable, so the two measurements of waste quantity can be compared. Variations exist primarily from moisture loss
during refuse storage and evaporation. Because the degree to which these mass reductions
occur within the MWC facility is uncertain, billing measurements should use the truck
scale. Clearly, feed rate measurements used in process analysis, however, should use the
crane load cells to more accurately reflect the waste burned. Information from both sources
should be transmitted to the central control room for use by operators in monitoring the
waste brought to the facility as well as the quantity being charged.
2.

Monitoring of Refuse Feed Rate

Refuse feed cranes are generally controlled manually by the operator who guides the
bridge and trolley travel to pick up loads and centers the grapple over the feed hopper,
raises the grapple and releases it. These control functions should be specified in the crane
manufacturers scope of supply. For plants in excess of 500 tpd, it may also be desirable to
have the crane controls included and automatic homing control to bring the grapple from
its location in the pit to directly over the feed chute. This saves time and also reduces
operator fatigue.
3.

Combustion Process Control

Combustion control can be complex, but maximum throughput and compliance with
regulations are facilitated by sophisticated control equipment and computerized control.
Large-scale mass burn incineration systems typically include vendor proprietary software
for automatic combustion controls. For the larger, energy recovery plants, the steaming

rate is used as the central control parameter. Some minimum automatic control is
necessary to ensure steady operations and environmental compliance and to protect
personnel and equipment. The combustion process itself requires a combination of
manually adjusted open-loop controls and automatic controls to maintain steady operations.
a. Manual Controls. The following combustion process parameters and functions can be
manually controlled either in the field or through open-loop control from the facility
control room.
FIELD-CONTROLLED PARAMETERS

Underfire air plenum damper settings. The underfire air distribution along the grate should
not require frequent adjustment. The dampers can be set manually during startup and
commissioning of the unit for optimum airflow to the drying, combustion, and burnout
zones. Thereafter, damper adjustments may be adjusted periodically as needed to react to
changing process conditions.
Overfire air distribution. As with the underfire air distribution, overfire air distribution can
be set manually during startup and commissioning of the unit to achieve the appropriate
mixing and combustion burnout and then checked periodically. Overfire air distribution
should be set by monitoring the CO and O2 concentrations and combustion temperatures
and by visual observation. If possible, the spatial distribution of CO and O2 should be
measured as an indirect measure of the mixing effectiveness of the overfire air jet system.
If the capability is built into the design, the direction of the overfire air jets can be modified
to more effectively mix the gases rising from the fuel bed.
Steam jet distribution. Although not practiced to a great extent in MWC systems, steam
nozzles have been used extensively to provide a means of mixing the combustion gases
with minimum mass addition. As for overfire air injection, the optimum steam
distribution does not require frequent adjustment. Therefore, manual control valves can
be provided to properly adjust the system during startup and commissioning of the MWC
system.
Grate speed (and the frequency of the feed ram if incorporated) should have local control
stations for field operation and maintenance. However, the control logic for these systems
is often integrated and controlled from the central control room during routine operations.
Siftings removal can be continuous or can be initiated from a field control station.
Generally, control from the central control room is not necessary. Indication that the
system is in use should be displayed in the control room.
Boilers are equipped with a safety valve and a blow-off valve. The latter is set manually at
the maximum safe pressure and releases when excess steam is being produced.
OPEN LOOP CONTROLS

Grate speed. Adjustment of grate speed depends on the feed rate and consequent heat
release rate of refuse and the quality of the ash product. The waste processing rate, and
therefore the grate speed, can be under manual control or can be tied to boiler steaming
rate. If the heating value of refuse is found to vary significantly during operations and the
situation cannot be controlled by mixing of waste in the pit, grate speed may be left to
operator judgment based on observation of the waste and knowledge of the chamber
conditions. The control of the feed ram should be made integral with the grate speed
controls such that this operation is properly sequenced during a feed cycle.

Underfire air supply. The flow rate of underfire air can be manually controlled by the
operator in the control room. This may be accomplished by using a variable-speed fan or
by using a damper to control flow. The operator adjusts undergrate air flow in response to
observations of the primary combustion air temperature, flue gas oxygen content, and
underfire air plenum pressure.
Overfire air supply. The overfire air distribution can be manually set to achieve a discharge
velocity that ensures penetration of the air jets across the combustion chamber through
adjustment of dampers. The flow rate of overfire air can be manually controlled by the
operator in the control room. This may be accomplished with a variable-speed fan or a
damper to control flow in consideration of the flue gas oxygen content, CO levels, and
furnace exit temperature.
b. Automatic Controls. To maintain steady operation and ensure environmental compliance, automatic control is often preferred. Parameters that benefit from automatic control
include the following.
Furnace draft control: The products of combustion, water vapor and excess air within the
primary furnace are removed by the induced draft (ID) fan. The ID fan pulls the gases from
the furnace through the various downstream components to maintain a constant draft
within the furnace at all times to prevent the escape of pollutants (smoke emission into the
incinerator building) due to local positive pressures within the system. Closed-loop control
of furnace draft is required and can be readily accomplished with a pressure transmitter
coupled to the induced draft fan speed or inlet vane damper. As the air flow through the
fuel bed changes, induced draft is adjusted automatically to maintain a balanced draft
condition in the furnace. The set point for furnace draft can be adjusted manually. The
location of the draft gauge on which furnace draft is controlled should be selected to avoid
influences such as overfire air or steam jets. To minimize pulling tramp air into the system,
the draft should be set at the lowest level that reliably avoids smoking.
Afterburning zone temperature. Pollution control relating to the emission of incomplete
products of combustion (CO, dioxins and their precursors, and other hydrocarbon-based
pollutants) can be achieved by maintaining temperature and oxygen concentrations within
target limits in combination with turbulence in the gas stream and a minimum residence
time to allow for molecular-scale mixing. In many countries, temperature and residence
time parameters are set by regulation. Often, 1000! C and 2 sec after the flue gases are mixed
(the plane just above the overfire air injection nozzles) is set as the baseline for temperature
and residence time. If the temperature drops below the mandated level, it is often necessary
either to stop feeding waste or to increase the temperature by firing fossil fuel.
An auxiliary fuel burner is often mounted in the furnace walls ahead of the first
boiler pass. The burner maintains flue gas temperature during startup and in the event that
the flue gas temperature cannot be maintained via air adjustment due to low calorific value
of the waste and=or high-moisture content. Thermocouples ahead of the first boiler pass
should be used to automatically start the burner when the temperature falls below the set
point. Burner startup and flame safety controls should be provided as a package with the
burner by the vendor and integrated with the system.
Boiler drum level. If a boiler is incorporated into the process design, water levels within the
drum should be automatically controlled using level probes coupled to the feedwater
pumps or control valve.

Steam demand. For some steam applications, closed-loop control may be required. This
would especially be true when steam is used in a steam turbine or engine to drive pumps or
fans. In this case, the flow demand of the device controls a steam valve. The same would
also be true for space-heating applications where steam flow would be controlled from a
temperature set point. Diversion of unused steam to a condenser or blowdown to
atmosphere would be provided.
Back-end temperature. The boiler, air preheater, air pollution control (APC) system, and
induced draft fan are sensitive to temperature. In most plants, either a boiler or a water
spray system is used to reduce the flue gas temperature to be safe for the back-end
equipment. If permitted, a bypass dump stack should be provided for emergency
conditions where a failure in boiler-water supply or other malfunction leads to excessive
temperatures in the back-end hardware. If the gas temperature exceeds a preset safe level
that protects the equipment from damage, the dump stack damper would open automatically. Alternatively, an alarm can be provided to alert the operator to manually engage
the damper. The damper release concept should be developed in light of the plausible
failure modes affecting the facility to include failure of the water or electricity supply, etc.
System startup (optional). It may be desirable to include an automatic system startup
control sequence that uses the auxiliary burner to automatically ramp up the system
temperature at a constant rate prior to the feeding of refuse and automatically ramp the
system down during shutdown. This can minimize thermal stresses on the system due to
rapid thermal expansion and contraction. The startup sequence can be manually controlled,
however; operators will need to be careful not to heat the system too rapidly. Likewise
when cooling down the unit, the induced draft fan should be controlled to provide a steady
but controlled rate of cooling.
Emergency shutdown. Attention should be paid in the design of the control system to
include fail-safe and automatic controls to protect personnel and equipment in the event of
a significant process upset, malfunction, and water or power failure.
Alarms. Key parameters being measured should also have alarm capability to alert
operators to the problem so that the appropriate control measures can be taken. Alarms
can be in two tiers, with the first being simply to alert the operator to attend to the problem
and the second actually to trigger an automatic sequence to prevent injury to personnel or
equipment. Parameters that would need to be alarmed include
Grate drive not functioning (zero-speed switch)
Residue discharger or conveyor not functioning
Zero-speed switches or pressure transmitters on all fans to sense no-flow
Combustion chamber overtemperature
Low boiler-water level
High temperature at air pollution control device inlet
Excessive steam temperature or pressure
Emission limits for CO or SO2 exceeded, failure of carbon or lime slurry feed system
c. Control Logic. In the early years of waste incineration, the process control strategy
was entirely in the hands of the operator. He looked at the waste, the fire, the residue
conveyor, and the stack and set the air and grate speed in accord with an unwritten but
often effective algorithm developed through personal experience. As incineration evolved
from a fire in a box for volume reduction to a power plant generating high-temperature,

high-pressure superheated steam, and as the environmental requirements and abatement

systems have undergone revolutionary changes from the control of blackbirds (large
chunks of charred paper ash) using simple wet bottom expansion chambers to complex,
multitechnology abatement trains, the process control concepts have also matured.
Control of incineration systems started with a strategy based on visual observations
supported with a minimum of indicating instruments leading to manual adjustment of a
few basic process variables. In a modern plant, control is centered in a computer that draws
information from several real-time sensors testing a variety of system properties and, using
a sophisticated system algorithm, exercises direct, automatic control over a suite of seminal
process parameters.
Martin GmbH mass burn systems, for example, use flue gas temperature from the
second boiler pass, measured by an infrared pyrometer, to control the refuse feeder and the
grate. This flue gas temperature is processed in a cascade control loop with the boiler
steam flow in order to adjust for the boiler load. A second control loop monitors the O2
levels in the flue gas to adjust the underfire air rate per underfire air compartment in order
to maintain a constant excess rate. The dry scrubber system monitors stack SO2 and makes
real-time adjustments to the lime slurry feed rate to ensure compliance with the SO2
emission limits.
Another system vendor, Detroit Stoker Company, regulates its stoker-fired boilers by
adjusting fuel and combustion air from changes in steam pressure. A change in steam
demand initiates a signal from the steam-pressure controller, through the boiler master
controller, to increase or decrease both fuel and air simultaneously and in parallel to satisfy
the demand. As long as an error in pressure exists (a departure of actual pressure from the
set point value), the steam pressure controller will continue to integrate the fuel and air
until the pressure has returned to its set point.
A combustion guide, in this case steam flow and air flow, compares pressure error to
a calibrated air flow demand to modify the rate of fuel being burned (proportional to steam
flow) and initiate any required readjustments to the forced draft damper so that the desired
fuelair ratio is continuously maintained. Furnace draft is regulated separately through the
use of a furnace-draft controller and a power operator that positions the uptake damper on
the induced draft fan.
4.

Air Pollution Control Train

Control of the air pollution control (APC) system will be dependent on the type chosen.
However, the basic concepts are similar among systems. Most of the controls would be
automatic to ensure that emission limits are being met and would be typically be provided
within the APC system vendors scope. The complexity of the control system and vendorspecific design warrants that the vendor provide a complete system. As with other vendorsupplied systems, clear limits of the scope of supply as well as the interface issues should
be well defined.
a. Manual Controls. There are limited functions associated with the APC system that
should be controlled manually in the field or through open-loop control from the facility
control room. These include the following field-controlled and indicated parameters.
Reagent storage systems. Chemicals used in the APC process (such as lime, activated
carbon, and ammonia or urea) will require on-site storage in silos (dry reagents) or tanks
(liquids). Delivery and unloading include local operator stations to start and stop these
operations. Silos and tanks should be equipped with high-high and low-low alarms as
well as pressure-relief valves if they are enclosed.

Chemical preparation systems. Some chemicals may have to be prepared within the plant
in batch operations. These could include preparation of caustic solutions or lime slurry.
These batch operations can be supplied with local control panels for operation of mixers,
circulating pumps, and pH adjustment. Water addition can be manual or automatic. An
alarm should be included to indicate low tank levels to alert operators to make new
batches.
Pressure and temperature. Each component of the APC train should be monitored for inlet
and outlet temperatures, inlet and outlet draft, as well as draft loss. Measurements are
displayed locally on the equipment, transmitted to the APC panel for use in controls, as
well as monitored and alarmed from the control room.
Activated carbon injection system. Depending on the emission limits for the plant and the
characteristics of the waste stream, combustion performance, etc., an activated carbon
injection system may be employed for control of mercury and dioxins. The feed rate of
carbon is empirically established based on waste composition as confirmed through the
stack testing of the system. Once set, the feed rate should be constant and not require
adjustment or control.
b. Automatic Controls. Several areas of the APC process are best controlled automatically to ensure that emissions limitations are being met and to maintain steady operation.
Baghouse inlet temperature control. Proper control of the inlet flue gas temperature is
needed to ensure that the baghouse does not experience elevated temperatures which could
cause serious damage to the collection bags. A thermocouple at the exit of the water spray
tempering system or the spray dryer outlet temperature thermocouple provides a feedback
signal to a temperature-indicating controller. The output from the controller is used as a set
point for a flow-indicating controller. This controller modulates the cooling water control
valve based on a feedback signal from a flow meter. The cascade loop automatically
adjusts for changes in water pressure without requiring a change in the outlet temperature.
Acid gas control. The flow rate of lime addition to the system is controlled by a SO2
monitor in the stack that measures the concentration. The SO2 monitor is used to control
the lime feed pumps or control valve.
Fabric filter cleaning control. The differential pressure across the fabric filter is continuously monitored. Reaching the set point triggers bag cleaning cycles. For pulse-jet
systems, compressed air is periodically pulse-injected through individual compartments
containing rows of bags. Each compartment is automatically brought off-line during the
cleaning process to prevent re-entrainment of dust. The cleaning cycle can be defined by a
timer or triggered by the pressure drop increase across the fabric filters. In addition, the
pulse may be manually initiated. The sequencing control automatically selects specific
compartments for cleaning once the cycle is initiated.
For shaker systems, the cleaning cycle will be set by the maximum pressure drop
allowed with extra compartments provided for off-line cleaning. For reverse-air baghouses,
the compartment to be cleaned is brought off-line and reverse flow initiated to flex or
collapse the filter bags in order to release the cake.
Fabric filter temperature. An automatic bypass damper can be provided to divert hot flue
gas and protect the filter bags in the event of excess temperature.
Dust level control. The fabric filter and spray dryer normally have rotary airlock or double
dump valve air seals to remove collected particulate. Automatic control is recommended
although these devices can be manually controlled. They can be operated on a timed
sequence. The timing sequence may be manually adjusted to suit the operating conditions.

A hopper level indicator should be provided to set of an alarm if a rotary or double dump
valve fails to operate properly and discharge material. This alarm function can also be used
to activate a vibrator to automatically dislodge material or to simply alert the operator.
Continuous-emission monitoring. Many jurisdictions require continuous-emission monitoring (CEM) for pollutants such as CO, O2 , SO2 , NOx ; and dust (opacity). CEM sampling
and analysis instruments are highly sensitive and require constant operator attention to
perform calibrations and maintenance. Control of the instruments is automatic and should
be included under the CEMs vendor scope of supply. To reduce equipment redundancy,
these instruments should provide the signals necessary for operator control and monitoring
of CO and O2 as well as for automatic control of acid gases.
Where permit requirements specify a minimum percent removal rate, additional
instruments may be required for measurement of the uncontrolled emissions prior to the air
pollution control system for automatic calculation of the removal efficiency. The CEM
systems should also be provided complete with a data logging and report system (usually
PC-based). In addition, instances where emission limits are exceeded should be alarmed to
the control room to alert the operations staff.
Emergency shutdown. The APC system should include fail-safe and automatic controls to
protect personnel and equipment in the event of a significant process upset, malfunction, or
power failure.
Alarms. Alarm capability is needed to alert operators to problems, triggering prompt
initiation of control measures. Commonly, alarms are set up in two tiers, with the first
alerting the operator to attend to the problem and the second triggering an automatic
sequence to prevent injury to personnel or equipment. Parameters that should be alarmed
include
Low reagent (lime, carbon, ammonia=urea, etc.) levels.
Water=slurry pumps on=off.
Spray dryer excess temperature (at fabric filter inlet).
Emission exceeds permit limits.
High level in dust collection hoppers.
5.

Ash Handling System

The ash handling systems can be manually controlled from either a local control station or
the central control room. In any case, indication of the system operating should be
observable (perhaps with TV monitors) from the control room.
a. Bottom Ash Collection System. Discharge of bottom ash from the combustion unit has
been described under the incinerator section above and would occur automatically from the
refuse feed control loop. Siftings removal, however, should be either continuous or
manually controlled from the local control station. Indication that the system is functioning
should be made to the central control room. If bottom ash and siftings are discharged to a
container system, no control is needed. However, if continuous conveyors are used, the
following guidelines should be considered.
The bottom ash conveyors should operate continuously to avoid buildup. Bottom ash
will be continuously discharged from the combustion unit.

If multiple conveyors are used, they should be coupled together to start in sequence
with the conveyor at the discharge end starting and stopping first, followed by the
next conveyor upstream. The conveyor at the incinerator discharge should be the
last to start in the sequence.
Conveyors should be equipped with zero-speed switches on the drive shaft to
indicate malfunctions to the control room.
Drives and motors should be placed at the head (discharge end) of the conveyor.
b. Fly-Ash Collection System. Regulations may mandate separation or allow combination
of fly ash and bottom ash. When separate handling is required, fly ash is usually collected
and stored in silos. Conveyance of fly ash can involve screw conveyors (up to no more than
30! in incline) or bucket conveyors (for steeper slopes) or can use pneumatic systems to
convey the ash to storage silos. The following guidelines should be considered:
The fly-ash conveyors should operate continuously to avoid buildup in the
conveyors.
If multiple conveyors are used, they should be coupled together to start in sequence
with the conveyor at the discharge end starting and stopping first followed by the
next and so forth up to the point where ash is loaded from the collection hoppers.
Conveyors should be equipped with zero-speed switches on the drive shaft to
indicate malfunctions to the control room.
Drives and motors should be placed at the head (discharge end) of the conveyor.
Level controls and alarms as well as pressure-relief valves should be provided on
storage silos.
Ash conditioning devices and silo discharge mechanisms should have a local control
station.
Indication that the system is in use should be provided in the control room.
6.

Auxiliaries

Auxiliary systems should be provided with local control stations and=or panels as
warranted by the design. Key operating parameters and functions should be indicated
and alarmed to the central control room. Examples of auxiliary systems that may be
incorporated in the plant include
Fire pumps
Refuse bunker drain pumps
Refuse dewatering device
Refuse bunker motorized delivery doors
Packaged wastewater treatment plant
Feedwater treatment system

E.

Portable Instruments

The operating staff should have access to portable instruments to make settings that do not
require frequent adjustments or to double-check on electronic measurements. Examples
include portable pressure gauges that allow manual settings of individual underfire air

plenums, portable flue gas analyzers to check on a permanently installed devices, and
hand-held temperature instruments.
F.

Summary

Table 17 summarizes essential measurements and controls described above. While the
control room operator has access to all measured variables and can override all control
loops, the instruments and controls required by the floor operator are mounted on a local
panel or control station as shown in the table. This table is a guideline only and should be
reviewed and modified for the particular requirements of a project and the types of
equipment and systems employed in the design.

XVIII.

OPERATIONS

A comprehensive treatment of the operating problems and challenges of municipal waste

incineration would introduce many new topics. Further, the rightness or wrongness of
many of the recommendations would be subject to controversy since the human element,
leadership styles, cultural effects, and the like often produce many good and workable
answers to the same question. However, there is merit is identifying the major operating
problems observed in existing plants as indications to the design engineer that attention is
necessary.
A.

Mass Burn Incineration

A compilation of such experiences based on a study of the equipment of approximately 30

manufacturers in 52 mass burn incineration plants (384) provides a useful data base. The
results of the survey are summarized in Table 18.
Refractory problems were the most common, ranging from the need for minor
patching to complete replacement. Minimization of the problem included both attention to
avoid charging bulky, metal wastes and replacement of (softer) castable refractory with the
more abrasion-resistant, fired brick.
Underfire air ports in modular combustion units were subject to plugging. Enlarging
of the orifices and periodic steam purging have been tried to help the problem, but regular
maintenance and cleaning are still necessary.
Many plants complain about the limitations of a small tipping floor, which causes
congestion during deliveries and does not give room for floor-dumping to inspect
incoming waste for auto batteries, hazardous wastes, etc., which local authorities may
require. Also, complaints called for the designers to give more attention to the potential
traffic flow pattern on the tipping floor.
Warping of dampers and charging doors was also common, especially in starved-air
systems and when units are run at higher than design temperatures. Similar problems with
charging rams were noted although deficiencies in the hydraulic systems were the prime
source of the difficulties.
B.

RDF Incineration

An analysis was prepared (533) of operational problems in a three-furnace, 2700-tonper-day (waste receipt) RDF incineration plant using shred and burn fuel preparation

Refuse scale
Refuse cranes
Bridge position
Trolley position
Refuse scale
Position
Weight
Primary chamber
Grate speed
Feed ram
Underfire air flow rate
Overfire air flow rate
Furnace temperature
Furnace exit temperature
Forced draft fan
Furnace draft (pressure)
Underfire air temperature
Underfire air flow
Oxygen monitor
CO monitor
Burner status
Boiler
Steam pressure
Saturated steam temp.
Superheated steam temp.
Steam flow
Drum level
Gas temp. at screen tubes
Temp. at economizer inlet
Temp. at economizer outlet

Description

Local panel=board

Field

X
X

X
X
X
X
X
X
X
X
X

X
X

X
X
X
X
X

X
X

S
S
F=P
F=P
T
T
F
P
T
F=P
A
A

P
T
T
F
L
T
T
T

X
X

X
X

X
X
X
X
X

X
X

X
X

X
X
X

X
X
X
X
X
X
X
X
X

X
X

X
X

X
X
X

X
X
X
X

X
X

X
X

X
X
X
X
X
X

X
X

X
X
X
X
X

X
X
X
X

X
X

X
X

X
X
X

X
X
X
X
X
X
X
X
X
X

X
X

Closed- OpenClosed- OpenClosed- Openloop

loop
Manual Indica- Record- loop
loop
Indica- Record- loop
Indica- Record- loop
control control Alarm Trip
tion
ing
control control Alarm Trip control
ing
control control Alarm Trip tion
Status tion
ing

Central control room

Table 17 Alternatives in Instrumentation and Control

Key: F Flow
T Temperature
P Pressure

X
X

X
X
X
X

F
F
T
F

X
X
X

S
S

L
S
S
L

X
X
X
X
X
X
X
X

L
F
L
S
T
P
P
A
A
A
F
X
X
X
X

X
X
X
X
X
X

L Level
W Weight
A Analysis
S Speed=frequency

Air pollution control

Lime reagent storage
Quench water=slurry flow
Carbon reagent storage
Carbon feed system
Spray dryer exit temp.
Spray dryer exit pressure
Fabric filter cleaning cycle
Opacity monitor
SO2 monitor
Other CEMS
Flue gas flow
Bottom ash handling
Siftings conveyor
Ash discharger
Bottom ash conveyor
Fly-ash handling
Dust hopper level
Rotary=double dump valves
Screw conveyors
Ash silo level
Auxiliaries
Storage pit drain pump
Boiler feedwater treatment
Boiler feedwater temp.
Wastewater treatment

Description

Local panel=board

Field

X
X

X
X
X

X
X
X
X

X
X
X
X

X
X

X
X
X
X

X
X
X
X
X

X
X

X
X
X
X
X
X
X
X
X
X

X
X
X

X
X

X
X
X
X

X
X
X
X
X
X

X
X
X

X
X

X
X
X

X
X

X
X
X

X
X
X
X

X
X
X

Closed- OpenClosed- OpenClosed- Openloop

Indica- Record- loop
Indica- Record- loop
loop
loop
Manual Indica- Record- loop
control control Alarm Trip
tion
ing
control control Alarm Trip control
ing
control control Alarm Trip tion
Status tion
ing

Central control room

Table 17 Alternatives in Instrumentation and Control (Continued)

Table
Problem
Refractory
Tipping floor
Fire tubes
Ash conveyor
Controls
Water tubes
Steam demand
Feed hopper
Stack damper
Front end loader

Recovery
Percent with
problem
71%
29%
25%
23%
19%
17%
13%
10%
4%
2%

Problem

Percent with
problem

Underfire air ports

Charging ram
Air pollution
Not on-line
Waste supply
Internal ram
I.D. fans
Quench water
Charging grate

36%
25%
23%
21%
19%
18%
12%
8%
2%

Source: From (385).

Table 19 RDF Incinerator Plant Availabilities

1991 Lost time
distribution
Category

Description

Scheduled

Hours

Subtotals

Tube overlays
Weeky maintenance
Boiler wash
Yearly maintenance
Forced
Feed systeme
Grate
Tube leake
Misc. trips
Bottom ash systeme
Equipment=Fuel Back pressuree
Opacitye
Wet fuele
Fuel supply
Force Majeure
Storms

352.0 1242.8
253.0
52.4%
349.8
315.0
121.7 829.9
248.1
35.5
369.6
84.0
6.5
93.1 250.1
32.6
10.6%
3.0
121.4
47.0
47.0
2.0%
Totals 2369.8

1992 Lost time

distribution
Hours

Subtotals

507.8 1385.3
140.0
65.0%
511.5
226.0
40.9 627.9
216.0
29.5%
329.5
35.5
6.0
38.7 112.6
16.3
5.3%
12.6
45.0
4.0
4.0
0.2%
2129.8

1993 Lost time

distribution
Hours Subtotals
182.0 904.0
142.0 58.9%
443.5
136.5
83.0 603.7
307.7 39.3%
133.8
79.2
0.0
5.8 24.5
8.2
1.6%
6.6
3.9
2.0
2.0
0.1%
1534.2

Summary Statistics
Fuel processing capacity loss
Boiler availabilityb
Feed system availabilityc
Plant availabilityd
a

8.30%
90.31%
97.86%
87.90%

5.30%
89.57%
98.97%
88.52%

7.70%
90.21%
98.28%
88.48%

Fuel capacity loss based on fuel handling facilities lost hours.

Boiler availability based on scheduled lost hours for Inconel tube re-cladding, quarterly boiler wash and yearly
maintenance on electrical equipment plus forced lost hours due to grate failures (esp. link breaks), tube leaks, and bottom
ash conveyor problems.
c
Feed system availability is calculated from combined lost hours of preventive maintenance and feed system problems.
d
Plant availability includes force majeure in addition to the hours assigned to the boiler and feed systems.
e
Other: Feed systemjams from stringy material, drag chain failure; tube leaksespecially from sootblower erosion;
bottom ash systemclinker; back pressureforced reduction in feed rate when condenser pressure is too high;
opacityfeed reduction when opacity>limit.; wet fueljams in feed system.
Source: From (533).
b

(single-stage shredding to about 15 cm followed by a single, magnetic separation stage

removing up to 60% of the ferrous). Each boiler has its own spray drier-absorber and either
a 5-stage ESP (boilers 1 and 2) or a 12-compartment baghouse (boiler 3). Urea injection is
used for NOx control. The 380 ton=hr of steam generates 80 MW of electricity using two
turbine=generators exhausting to dry, air-cooled condensers. Table 19 summarizes the
types of scheduled and forced outages the plant experiences.